LJungang/Awesome-Video-Reasoning-Landscape

🔥An open-source survey of the latest video reasoning tasks, paradigms, and benchmarks.

Python

191

148 commits

updated Sep 24, 2026

See the code

README

Video Reasoning Landscape

🗺️ At a Glance

A curated survey of reasoning over videos, reasoning through visual generation, and building worlds for prediction and action.

The landscape has two complementary views: reasoning engines—text, frames, structured/latent states, and programs—and interaction loops—interleaving and streaming. A system can combine several engines within one loop.

Browse the Paper Atlas, choose a benchmark, or contribute a paper.

📖 Contents

🧭 Start Here
📚 Paper Atlas
🎯 Choose Your Benchmark
🔬 Evaluation Playbook
🔭 Next Frontiers
📐 Modeling Notes
✈️ Further Reading
🤝 Contributing
🌟 Star History
♥️ Contributors

🧭 Start Here: Choose Your Research Track

Your question	Read	Compare on
Can a model answer from the right video evidence?	Grounded Reasoning · Stateful Intelligence	CaST-Bench, SagaQA: answer correctness and evidence localization
Does generating a visual trajectory help solve a problem?	Thinking in Visual Futures	VBVR-Pro, MME-CoF-Pro: intermediate rule validity and final success
Can a model construct and execute a consistent world?	Executable Worlds	WorldCoder-Bench: runtime state, physical constraints, and interaction contracts
Can an assistant search, remember, and respond at the right time?	Active Video Agents · Always-On Intelligence	StreamArena, ProactiveBench: observed-prefix access, memory, and response timing
Do imagined futures improve action selection?	World Models for Action	WorldArena: rollout fidelity and downstream decision utility

Mechanisms and access. Language traces, generated frames, structured memory, and executable programs are different intermediate representations. Tool use and streaming describe how a system accesses evidence; either can wrap several representations. RL and distillation describe training, not a separate reasoning mechanism.

Read world-model claims by component. GraphThinker organizes observed state; Code World Model separates executable state from learned rendering; Code-as-World searches for executable explanations of observations. Evaluate state estimation, transition rules, and rendering separately—one score cannot establish all three.

For a comparison protocol, use Choose Your Benchmark and the Evaluation Playbook. Modeling Notes defines the evidence loop and hybrid world interface; Worlds as Programs examines their scientific limits.

📚 Paper Atlas

Time is the first-public month; (proc.) marks a proceedings date with an unresolved earlier preprint. N/A means unrecorded or not applicable.

Input modalities & benchmark tasks

Badges describe supplied inputs, including optional conditioning. They exclude outputs, internal representations, and task names. Lists cover verified inputs and may be incomplete.

Modality	Input	Color
![Text][mod-text]	Questions, instructions, system prompts, transcripts, or textual context	`#AEC6DF`
![Video][mod-video]	Observed clips, streams, or temporally ordered frames	`#C3E6CB`
![Audio][mod-audio]	Waveforms, speech, or environmental sound	`#F7D6AE`
![Image][mod-image]	Still images or reference images	`#D7C4E8`
![Depth][mod-depth]	Depth maps or RGB-D observations	`#F2C6D6`
![Point Cloud][mod-point-cloud]	Observed 3D point sets	`#B9E3E0`
![Action][mod-action]	Control commands or action sequences supplied to a model	`#F6E3A9`
![State][mod-state]	Supplied object, environment, or robot states and structured annotations	`#D8DEE9`
![Code][mod-code]	Source code or programs supplied as input	`#C9C9F1`
![Mesh][mod-mesh]	Supplied 3D geometry or animated meshes	`#E0D0B8`

Benchmark Task identifies the evaluated model family, not its input modality. Both badges indicate evaluation of both families; N/A means unspecified or outside these families.

Task	Evaluated reasoning	Color
![language][task-language]	Reasoning by language models, including multimodal LMs	`#AEC6DF`
![vision][task-vision]	Reasoning by visual generation models	`#C3E6CB`

🧠 Reasoning Engines

🗒️ Grounded Reasoning

Language-led inference, evidence-linked training, distillation, and reliability.

Paper	Resources	Input modalities	Time	Venue
What Do Hallucinations Reveal About Multimodal Reasoning? Diagnosing Visual Grounding Failures via Contrastive Decoding Probes	![resource:code][res-github-code]	![Text][mod-text]	2026-09	`arXiv`
Efficient Reasoning Distillation: Small Video-Language Models via Synthetic CoT and Difficulty-Aware Fine-Tuning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Long-to-Short Video Evidence Reasoning for Grounded Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
BodyCam-VQA: Enhanced Body-Worn Camera Video Captioning via Multimodal Reasoning and Probe Question Generation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Video-MOPD: Multi-Teacher On-Policy Distillation for Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Learning Compositional Spatio-Temporal Video Grounding with Synthetic Curriculum	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Video-OPSD: Exploiting Privileged Visual Evidence for On-Policy Self-Distillation in Video Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Self-Reflective Multi-modal Reasoning for Short-Video Fake News Detection	`N/A`	![Text][mod-text]	2026-08	`arXiv`
Reason in the Words You Speak: Idiolectal Paraphrasing Off-Policy Traces for Reasoning Distillation in VideoLLMs	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Video-FLAIR: Not Whether to Reason, But How	`N/A`	![Text][mod-text]	2026-08	`arXiv`
Finding the Right Evidence: Factor-Guided Coarse-to-Fine Reasoning for Long Videos	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Where to Look Matters: On-Policy Self-Distillation for Long-Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Multi-Agent Self-Improving Reinforcement Learning for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Improving Spatial-Temporal Reasoning in Video-Language Models with Structured Video Prompting	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
COMET: Contrastive Motion-Enhanced Temporal Reasoning for Video Multimodal Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Enhancing Localized Reasoning for Long Video Understanding via Efficient Segment-to-Video Supervision	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Annotations as Rollouts: Efficient and Scalable Reinforcement Learning for Video MLLMs	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Deep Thought Alignment: Trajectory-Level Latent Distillation for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
TennisVAR: A Stroke-Evidence-Grounded Multimodal Large Language Model for Tactical Reasoning in Tennis Videos	![resource:project][res-project]	![Text][mod-text]	2026-08	`arXiv`
AdaThinkV: Adaptive Thinking for Token-Efficient Video Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Adaptive Emotional Video Captioning via Affective Heterogeneous Graph Reasoning and Multi-task Joint Learning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Knowledge-Guided Multimodal Reasoning over Interacting Streams for Video-Level Ambivalence and Hesitancy Recognition	`N/A`	![Text][mod-text]	2026-07	`arXiv`
CADER: Confidence-Aware Dynamic Evidence Reasoning for Long-Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
O-VAD: Industrial Video Anomaly Detection through Object-Centric Tracking and Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
ConsiSpace: Learning Geometric Consistency Matters for Video Spatial Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
GHR-VLM: Making Zero-Shot Transit Video Analytics Realizable with Grounded Hybrid Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Evidence-Backed Video Question Answering	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
TimeThink: Reasoning with Time for Video LLMs	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
EventCoT: Event-centric Video Chain-of-thought for Reasoning Temporal Localization	`N/A`	![Text][mod-text]	2026-07	`arXiv`
STAC: Selective Spatiotemporal Aggregation and Compression for Video Reasoning Segmentation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
EFlow: Learning Evidence Flow for Long-Video Reasoning with Adaptive Reflection	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Linguistic Relative Policy Optimization for Video Anomaly Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Reflect-R1: Evidence-Driven Reflection for Self-Correction in Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
SER: Learning to Ground Video Reasoning with Semantic Evidence Rewards	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
CineCap: Structured Reasoning with Spatio-Temporal Anchors for Cinematographic Video Captioning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
CARE: Competence-Aware Reward Shaping for Adaptive Reasoning Length in Video-MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
APT: Atomic Physical Transitions for Causal Video-Language Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Reasoning as Intersection: Consensus-Frame Alignment for Visual Focus in Video-MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Training LLMs with Reinforcement Learning over Digital Twin Representations for Reasoning-Intensive Surgical VideoQA	`N/A`	![Text][mod-text]	2026-06	`arXiv`
Temporal-Aware Reasoning Optimization for Video Temporal Grounding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Counterfactual Reasoning for Fine-Grained Evidence Disentanglement in VideoQA	`N/A`	![Text][mod-text]	2026-06	`arXiv`
See More, Think Deeper: Query-Expanded Visual Evidence and Answer-Clue Guided Reflection for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
When Video Misreads: Closed-Loop Distillation of Reading Heuristics for Exploratory Manipulation Trace QA	`N/A`	![Text][mod-text]	2026-06	`arXiv`
CACR:Reinforcing Temporal Answer Grounding in Instructional Video via Candidate-Aware Causal Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
VideoSEG-O3: A Multi-turn Reinforcement Learning Framework for Reasoning Video Object Segmentation	![resource:code][res-github-code]	![Text][mod-text]	2026-06	`arXiv`
VLMs are Good Teachers for Video Reasoning via Adaptive Test-Time Optimization	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Question-Aware Evidence Ledgers for Video Relational Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Training-Free Composed Video Retrieval via Visual Representation-Guided Video-LLM Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
TLG: Temporal-Logic Grounding for Video Question Answering via Source-Annotation Reconstruction and Category-Targeted Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video] ![State][mod-state]	2026-06	`arXiv`
Perception First: A Frontier Native-Video Model with Self-Consistency for Implicit Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
R^3: Composed Video Retrieval via Reasoning-Guided Recalling and Re-ranking	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Adaptive Dense Evidence Refinement for Video Relational Reasoning for VRR-QA Challenge	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Disagreement-Based Cross-Model Routing for Implicit Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Reason, Retrieve, Re-rank: A Zero-Shot Reasoning-Aware Framework for Composed Video Retrieval	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Rethinking Video-Language Model from the Language Input Perspective	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Reflective Dialogue between Teacher and Solver Agents for Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Clinically-Grounded Counterfactual Reasoning for Medical Video Diagnosis	`N/A`	![Text][mod-text]	2026-05	`CVPR 2026`
CoReVAD: A Contextual Reasoning Framework for Training-Free Video Anomaly Detection	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Learning Spatiotemporal Sensitivity in Video LLMs via Counterfactual Reinforcement Learning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Foresee-to-Ground: From Predictive Temporal Perception to Evidence-Driven Reasoning for Video Temporal Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
EvoVid: Temporal-Centric Self-Evolution for Video Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
SteerSeg: Attention Steering for Reasoning Video Segmentation	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Video-Zero: Self-Evolution Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Separate First, Fuse Later: Mitigating Cross-Modal Interference in Audio-Visual LLMs Reasoning with Modality-Specific Chain-of-Thought	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-05	`arXiv`
RCoT-Seg: Reinforced Chain-of-Thought for Video Reasoning and Segmentation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
VISD: Enhancing Video Reasoning via Structured Self-Distillation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Think, then Score: Decoupled Reasoning and Scoring for Video Reward Modeling	`N/A`	![Text][mod-text]	2026-05	`arXiv`
From Priors to Perception: Grounding Video-LLMs in Physical Reality	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Beyond Perceptual Shortcuts: Causal-Inspired Debiasing Optimization for Generalizable Video Reasoning in Lightweight MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`CVPR 2026`
Co-Evolving Policy Distillation	`N/A`	![Text][mod-text]	2026-04	`arXiv`
StoryTR: Narrative-Centric Video Temporal Retrieval with Theory of Mind Reasoning	`N/A`	![Text][mod-text]	2026-04	`arXiv`
UpstreamQA: A Modular Framework for Explicit Reasoning on Video Question Answering Tasks	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
ProcessThinker: Enhancing Multi-modal Large Language Models Reasoning via Rollout-based Process Reward	`N/A`	![Text][mod-text]	2026-04	`arXiv`
Seeing Fast and Slow: Learning the Flow of Time in Videos	![resource:project][res-project]	![Text][mod-text]	2026-04	`arXiv`
EasyVideoR1: Easier RL for Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Find, Fix, Reason: Context Repair for Video Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Lost in Adaptation: Layer-Selective Recovery of Temporal Reasoning in Video-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Do Thought Streams Matter? Evaluating Reasoning in Gemini Vision-Language Models for Video Scene Understanding	`N/A`	![Text][mod-text]	2026-04	`arXiv`
OmniJigsaw: Enhancing Omni-Modal Reasoning via Modality-Orchestrated Reordering	![resource:project][res-project]	![Text][mod-text] ![Audio][mod-audio]	2026-04	`arXiv`
Reasoning-Guided Grounding: Elevating Video Anomaly Detection through Multimodal Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
STEER: Structured Event Evidence for Video Reasoning via Multi-Objective Reinforcement Learning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Reinforce to Learn, Elect to Reason: A Dual Paradigm for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`CVPR 2026`
STEAR: Layer-Aware Spatiotemporal Evidence Intervention for Hallucination Mitigation in Video Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
STRIVE: Structured Spatiotemporal Exploration for Reinforcement Learning in Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Reinforcing Consistency in Video MLLMs with Structured Rewards	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
TTA-Vid: Generalized Test-Time Adaptation for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement Learning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Incentivizing Temporal-Awareness in Egocentric Video Understanding Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
VIRST: Video-Instructed Reasoning Assistant for SpatioTemporal Segmentation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
Reinforcing Structured Chain-of-Thought for Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
GridVAD: Open-Set Video Anomaly Detection via Spatial Reasoning over Stratified Frame Grids	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Video-Only ToM: Enhancing Theory of Mind in Multimodal Large Language Models	![resource:project][res-project]	![Text][mod-text]	2026-03	`CVPR 2026`
Learning Trajectory-Aware Multimodal Large Language Models for Video Reasoning Segmentation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
CoVR-R:Reason-Aware Composed Video Retrieval	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Narrative Aligned Long Form Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Learning Transferable Temporal Primitives for Video Reasoning via Synthetic Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
When Thinking Hurts: Mitigating Visual Forgetting in Video Reasoning via Frame Repetition	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Video-CoE: Reinforcing Video Event Prediction via Chain of Events	`N/A`	![Text][mod-text]	2026-03	`CVPR 2026`
VisionCoach: Reinforcing Grounded Video Reasoning via Visual-Perception Prompting	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
SPARROW: Learning Spatial Precision and Temporal Referential Consistency in Pixel-Grounded Video MLLMs	![resource:project][res-project] ![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
Beyond Single-Sample: Reliable Multi-Sample Distillation for Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Are Video Reasoning Models Ready to Go Outside?	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Video-Based Reward Modeling for Computer-Use Agents	`N/A`	![Text][mod-text]	2026-03	`arXiv`
Geometry-Aware Semantic Reasoning for Training Free Video Anomaly Detection	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
SarcasmMiner: A Dual-Track Post-Training Framework for Robust Audio-Visual Sarcasm Reasoning	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-03	`arXiv`
Weakly Supervised Video Anomaly Detection with Anomaly-Connected Components and Intention Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`CVPR 2026`
APPO: Attention-guided Perception Policy Optimization for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`CVPR 2026`
Think with Grounding: Curriculum Reinforced Reasoning with Video Grounding for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
STVG-R1: Incentivizing Instance-Level Reasoning and Grounding in Videos via Reinforcement Learning	`N/A`	![Text][mod-text]	2026-02	`arXiv`
Credit Where It is Due: Cross-Modality Connectivity Drives Precise Reinforcement Learning for MLLM Reasoning	`N/A`	![Text][mod-text]	2026-02	`arXiv`
VideoVeritas: AI-Generated Video Detection via Perception Pretext Reinforcement Learning	![resource:code][res-github-code]	![Text][mod-text]	2026-02	`arXiv`
Process-of-Thought Reasoning for Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
OmniVideo-R1: Reinforcing Audio-visual Reasoning with Query Intention and Modality Attention	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-02	`arXiv`
AVERE: Improving Audiovisual Emotion Reasoning with Preference Optimization	![resource:project][res-project]	![Text][mod-text] ![Audio][mod-audio]	2026-02	`arXiv`
GuardReasoner-Omni: A Reasoning-based Multi-modal Guardrail for Text, Image, Video, and Audio	`N/A`	![Text][mod-text]	2026-02	`arXiv`
Video-OPD: Efficient Post-Training of Multimodal Large Language Models for Temporal Video Grounding via On-Policy Distillation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
RANKVIDEO: Reasoning Reranking for Text-to-Video Retrieval	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
LongVPO: From Anchored Cues to Self-Reasoning for Long-Form Video Preference Optimization	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
SRVAU-R1: Enhancing Video Anomaly Understanding via Reflection-Aware Learning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Structure Over Scale: Learning Visual Reasoning from Pedagogical Video	`N/A`	![Text][mod-text]	2026-01	`arXiv`
Triage: Hierarchical Visual Budgeting for Efficient Video Reasoning in Vision-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Enhancing Video Representations with Spatiotemporal-Semantic Residual to Mitigate Hallucinations in Video Large Multimodal Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning	`N/A`	![Text][mod-text]	2026-01	`arXiv`
Video-KTR: Reinforcing Video Reasoning via Key Token Attribution	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Integrating Fine-Grained Audio-Visual Evidence for Robust Multimodal Emotion Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Audio][mod-audio]	2026-01	`arXiv`
Order from Chaos: Physical World Understanding from Glitchy Gameplay Videos	`N/A`	![Text][mod-text]	2026-01	`arXiv`
Training-Free and Interpretable Hateful Video Detection via Multi-stage Adversarial Reasoning	![resource:code][res-github-code]	![Text][mod-text]	2026-01	`arXiv`
Advancing Adaptive Multi-Stage Video Anomaly Reasoning: A Benchmark Dataset and Method	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
VERHallu: Evaluating and Mitigating Event Relation Hallucination in Video Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
CASHEW: Stabilizing Multimodal Reasoning via Iterative Trajectory Aggregation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Video Evidence to Reasoning Efficient Video Understanding via Explicit Evidence Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-01	`CVPR 2026`
CounterVid: Counterfactual Video Generation for Mitigating Action and Temporal Hallucinations in Video-Language Models	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Analyzing Reasoning Consistency in Large Multimodal Models under Cross-Modal Conflicts	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
PrismVAU: Prompt-Refined Inference System for Multimodal Video Anomaly Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
V-CORE: Temporally Consistent Video Understanding for Video-LLM	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Explicit Abstention Knobs for Predictable Reliability in Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
VideoCuRL: Video Curriculum Reinforcement Learning with Orthogonal Difficulty Decomposition	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`ACL 2026`
Robust Egocentric Referring Video Object Segmentation via Dual-Modal Causal Intervention	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
Factorized Learning for Temporally Grounded Video-Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
SmartSight: Mitigating Hallucination in Video-LLMs Without Compromising Video Understanding via Temporal Attention Collapse	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
Xiaomi MiMo-VL-Miloco Technical Report	![resource:code][res-github-code]	![Text][mod-text]	2025-12	`arXiv`
Rethinking Chain-of-Thought Reasoning for Videos	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
1+1 > 2 : Detector-Empowered Video Large Language Model for Spatio-Temporal Grounding and Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
TempR1: Improving Temporal Understanding of MLLMs via Temporal-Aware Multi-Task Reinforcement Learning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`
OneThinker: All-in-one Reasoning Model for Image and Video	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`
WorldMM: Dynamic Multimodal Memory Agent for Long Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`
Thinking with Drafts: Speculative Temporal Reasoning for Efficient Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-11	`CVPR 2026`
Video-R2: Reinforcing Consistent and Grounded Reasoning in Multimodal Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
Video-CoM: Interactive Video Reasoning via Chain of Manipulations	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
VideoSeg-R1: Reasoning Video Object Segmentation via Reinforcement Learning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
AVATAAR: Agentic Video Answering via Temporal Adaptive Alignment and Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-11	`arXiv`
Agentic Video Intelligence: A Flexible Framework for Advanced Video Exploration and Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
Video Spatial Reasoning with Object-Centric 3D Rollout	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
ViSS-R1: Self-Supervised Reinforcement Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
Video-Thinker: Sparking "Thinking with Videos" via Reinforcement Learning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-10	`arXiv`
Open-o3 Video: Grounded Video Reasoning with Explicit Spatio-Temporal Evidence	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-10	`arXiv`
MOSS-ChatV: Reinforcement Learning with Process Reasoning Reward for Video Temporal Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-09	`arXiv`
VideoChat-R1.5: Visual Test-Time Scaling to Reinforce Multimodal Reasoning by Iterative Perception	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-09	`arXiv`
Strefer: Empowering Video LLMs with Space-Time Referring and Reasoning via Synthetic Instruction Data	![resource:code][res-github-code] weights: Google_Drive	![Text][mod-text] ![Video][mod-video]	2025-09	`arXiv`
Kwai Keye-VL 1.5 Technical Report	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-09	`arXiv`
Video-MTR: Reinforced Multi-Turn Reasoning for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-08	`ICML 2026`
Ovis2.5 Technical Report	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-08	`arXiv`
Reinforcing Video Reasoning Segmentation to Think Before It Segments	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-08	`CVPR 2026`
TAR: Temporal Anchor-Constrained Reasoning for Video Temporal Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-08	`ECCV 2026`
ReasoningTrack: Chain-of-Thought Reasoning for Long-term Vision-Language Tracking	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-08	`arXiv`
Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video Reasoning	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-08	`arXiv`
AVATAR: Reinforcement Learning to See, Hear, and Reason Over Video	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-08	`CVPR 2026`
VideoForest: Person-Anchored Hierarchical Reasoning for Cross-Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-08	`ACM-MM 2025`
ReasonAct: Progressive Training for Fine-Grained Video Reasoning in Small Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-08	`arXiv`
ARC-Hunyuan-Video-7B: Structured Video Comprehension of Real-World Shorts	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-07	`arXiv`
METER: Multi-modal Evidence-based Thinking and Explainable Reasoning -- Algorithm and Benchmark	`N/A`	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-07	`arXiv`
CoTasks: Chain-of-Thought based Video Instruction Tuning Tasks	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-07	`arXiv`
EmbRACE-3K: Embodied Reasoning and Action in Complex Environments	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-07	`arXiv`
ViTCoT: Video-Text Interleaved Chain-of-Thought for Boosting Video Understanding in Large Language Models	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-07	`ACM-MM 2025`
Scaling RL to Long Videos	![resource:code][res-github-code] ![resource:data][res-hf-data] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-07	`NeurIPS 2025`
Video-RTS: Rethinking Reinforcement Learning and Test-Time Scaling for Efficient and Enhanced Video Reasoning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-07	`EMNLP 2025`
Kwai Keye-VL Technical Report	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-07	`arXiv`
Temporal Chain of Thought: Long-Video Understanding by Thinking in Frames	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-07	`arXiv`
DIVE: Deep-search Iterative Video Exploration	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`CVPR 2025`
HumanOmniV2: From Understanding to Omni-Modal Reasoning with Context	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-06	`arXiv`
VLN-R1: Vision-Language Navigation via Reinforcement Fine-Tuning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Ego-R1: Chain-of-Tool-Thought for Ultra-Long Egocentric Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
DAVID-XR1: Detecting AI-Generated Videos with Explainable Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
VideoDeepResearch: Long Video Understanding With Agentic Tool Using	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
VidBridge-R1: Bridging QA and Captioning for RL-based Video Understanding Models with Intermediate Proxy Tasks	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Wait, We Don't Need to "Wait"! Removing Thinking Tokens Improves Reasoning Efficiency	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
DeepVideo-R1: Video Reinforcement Fine-Tuning via Difficulty-aware Regressive GRPO	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`NeurIPS 2025`
VideoChat-A1: Thinking with Long Videos by Chain-of-Shot Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
MiMo-VL Technical Report	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Video-Skill-CoT: Skill-based Chain-of-Thoughts for Domain-Adaptive Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`EMNLP 2025 (Findings)`
EgoVLM: Policy Optimization for Egocentric Video Understanding	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Reinforcement Learning Tuning for VideoLLMs: Reward Design and Data Efficiency	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
VideoCap-R1: Enhancing MLLMs for Video Captioning via Structured Thinking	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
ReAgent-V: A Reward-Driven Multi-Agent Framework for Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`NeurIPS 2025`
ReFoCUS: Reinforcement-guided Frame Optimization for Contextual Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Video-CoT: A Comprehensive Dataset for Spatiotemporal Understanding of Videos Based on Chain-of-Thought	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Chain-of-Frames: Advancing Video Understanding in Multimodal LLMs via Frame-Aware Reasoning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`CVPR 2026`
SiLVR: A Simple Language-based Video Reasoning Framework	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`TMLR 2026`
Reinforcing Video Reasoning with Focused Thinking	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
Fostering Video Reasoning via Next-Event Prediction	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
A2Seek: Towards Reasoning-Centric Benchmark for Aerial Anomaly Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
Omni-R1: Reinforcement Learning for Omnimodal Reasoning via Two-System Collaboration	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-05	`NeurIPS 2025`
Vad-R1: Towards Video Anomaly Reasoning via Perception-to-Cognition Chain-of-Thought	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`NeurIPS 2025`
VerIPO: Cultivating Long Reasoning in Video-LLMs via Verifier-Guided Iterative Policy Optimization	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
Fact-R1: Towards Explainable Video Misinformation Detection with Deep Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-05	`NeurIPS 2025`
Pixel Reasoner: Incentivizing Pixel-Space Reasoning with Curiosity-Driven Reinforcement Learning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`NeurIPS 2025`
UniVG-R1: Reasoning Guided Universal Visual Grounding with Reinforcement Learning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
VideoRFT: Incentivizing Video Reasoning Capability in MLLMs via Reinforced Fine-Tuning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`NeurIPS 2025`
VISTA: Mitigating Semantic Inertia in Video-LLMs via Training-Free Dynamic Chain-of-Thought Routing	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
Seed1.5-VL Technical Report	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
TEMPURA: Temporal Event Masked Prediction and Understanding for Reasoning in Action	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
AVA: Towards Agentic Video Analytics with Vision Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`NSDI 2026`
MR. Video: "MapReduce" is the Principle for Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
Eagle 2.5: Boosting Long-Context Post-Training for Frontier Vision-Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
Multimodal Long Video Modeling Based on Temporal Dynamic Context	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-04	`arXiv`
TinyLLaVA-Video-R1: Towards Smaller LMMs for Video Reasoning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
LVC: A Lightweight Compression Framework for Enhancing VLMs in Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
Spatial-R1: Enhancing MLLMs in Video Spatial Reasoning	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
WikiVideo: Article Generation from Multiple Videos	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
Improved Visual-Spatial Reasoning via R1-Zero-Like Training	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
Aurelia: Test-time Reasoning Distillation in Audio-Visual LLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-03	`ICCV 2025`
Video-R1: Reinforcing Video Reasoning in MLLMs	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-03	`NeurIPS 2025`
VideoMind: A Chain-of-LoRA Agent for Temporal-Grounded Video Reasoning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-03	`ICLR 2026`
Time-R1: Post-Training Large Vision Language Model for Temporal Video Grounding	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-03	`NeurIPS 2025`
ST-Think: How Multimodal Large Language Models Reason About 4D Worlds from Ego-Centric Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-03	`NeurIPS 2025`
TheoremExplainAgent: Towards Video-based Multimodal Explanations for LLM Theorem Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-02	`ACL 2025 (Oral)`
video-SALMONN-o1: Reasoning-enhanced Audio-visual Large Language Model	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-02	`arXiv`
CoS: Chain-of-Shot Prompting for Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-02	`arXiv`
Temporal Preference Optimization for Long-Form Video Understanding	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-01	`arXiv`
InternLM-XComposer2.5-Reward: A Simple Yet Effective Multi-Modal Reward Model	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-01	`ACL 2025 (Findings)`
MECD+: Unlocking Event-Level Causal Graph Discovery for Video Reasoning	![resource:code][res-github-code] ![resource:weights][res-github-weights]	![Text][mod-text] ![Video][mod-video]	2025-01	`IEEE TPAMI`
Building a Mind Palace: Structuring Environment-Grounded Semantic Graphs for Effective Long Video Analysis with LLMs	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-01	`arXiv`
Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling	![resource:code][res-github-code] ![resource:weights][res-github-weights]	![Text][mod-text] ![Video][mod-video]	2024-12	`arXiv`
STEP: Enhancing Video-LLMs' Compositional Reasoning by Spatio-Temporal Graph-guided Self-Training	`N/A`	![Text][mod-text] ![Video][mod-video]	2024-11	`CVPR 2025`
VideoEspresso: A Large-Scale Chain-of-Thought Dataset for Fine-Grained Video Reasoning via Core Frame Selection	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2024-11	`CVPR 2025`
Adaptive Video Understanding Agent: Enhancing efficiency with dynamic frame sampling and feedback-driven reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2024-10	`NeurIPS 2024 (Workshop)`
VideoINSTA: Zero-shot Long Video Understanding via Informative Spatial-Temporal Reasoning with LLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2024-09	`EMNLP 2024 (Findings)`
MECD: Unlocking Multi-Event Causal Discovery in Video Reasoning	![resource:code][res-github-code] ![resource:weights][res-github-weights]	![Text][mod-text] ![Video][mod-video]	2024-09	`NeurIPS 2024 (Spotlight)`
Video-of-Thought: Step-by-Step Video Reasoning from Perception to Cognition	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2024-05	`ICML 2024`

🎞️ Thinking in Visual Futures

Generated frames and visual updates as intermediate computation, including image-based bridges. Scene time and diffusion steps are distinct.

Paper	Resources	Input modalities	Time	Venue
Do Video Generators Track the World Across Segments? A Benchmark and Method for World-State Reasoning in Video Continuation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Visual prompt engineering for video models	![resource:project][res-project]	![Video][mod-video]	2026-07	`arXiv`
Hierarchical Denoising For Multi-Step Visual Reasoning	![resource:project][res-project]	![Video][mod-video]	2026-07	`arXiv`
OpenCoF: Learning to Reason Through Video Generation	![resource:project][res-project]	![Video][mod-video]	2026-07	`arXiv`
DeltaV: Thinking with Visual State Updates in Unified Large Multimodal Models	![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
Temporal Backtracking Search for Test-time Generative Video Reasoning	`N/A`	![Video][mod-video]	2026-06	`arXiv`
Video Models Can Reason with Verifiable Rewards	`N/A`	![Text][mod-text]	2026-05	`arXiv`
Video Models Reason Early: Exploiting Plan Commitment for Maze Solving	`N/A`	`N/A`	2026-03	`arXiv`
Wan-R1: Verifiable-Reinforcement Learning for Video Reasoning	`N/A`	![Video][mod-video]	2026-03	`arXiv`
Demystifying Video Reasoning	`N/A`	![Video][mod-video]	2026-03	`arXiv`
A Very Big Video Reasoning Suite	![resource:project][res-project]	![Video][mod-video]	2026-02	`arXiv`
Thinking in Frames: How Visual Context and Test-Time Scaling Empower Video Reasoning	`N/A`	![Video][mod-video]	2026-01	`arXiv`
Visual Generation Unlocks Human-Like Reasoning through Multimodal World Models	![resource:project][res-project]	`N/A`	2026-01	`arXiv`
CoF-T2I: Video Models as Pure Visual Reasoners for Text-to-Image Generation	![resource:code][res-github-code]	![Text][mod-text]	2026-01	`arXiv`
VideoCoF: Unified Video Editing with Temporal Reasoner	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Video][mod-video]	2025-12	`CVPR 2026`
McSc: Motion-Corrective Preference Alignment for Video Generation with Self-Critic Hierarchical Reasoning	![resource:code][res-github-code]	`N/A`	2025-11	`arXiv`
In-Video Instructions: Visual Signals as Generative Control	![resource:code][res-github-code]	![Image][mod-image]	2025-11	`arXiv`
Video-as-Answer: Predict and Generate Next Video Event with Joint-GRPO	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text]	2025-11	`arXiv`
Reasoning via Video: The First Evaluation of Video Models’ Reasoning Abilities through Maze-Solving Tasks	![resource:code][res-github-code] ![resource:weights][res-hf-weights] ![resource:data][res-hf-data]	`N/A`	2025-11	`arXiv`
Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm	![resource:code][res-github-code]	`N/A`	2025-11	`arXiv`
Video Models Start to Solve Chess, Maze, Sudoku, Mental Rotation, and Raven’ Matrices	![resource:code][res-github-code]	`N/A`	2025-11	`arXiv`
Are Video Models Ready as Zero-Shot Reasoners? An Empirical Study with the MME-CoF Benchmark	![resource:code][res-github-code] ![resource:data][res-hf-data]	`N/A`	2025-10	`arXiv`
VChain: Chain-of-Visual-Thought for Reasoning in Video Generation	![resource:code][res-github-code]	`N/A`	2025-10	`ACL 2026 (Findings)`

🧩 Stateful Intelligence

Graphs, tracks, spatial maps, and latent reasoning states. Observed memory and imagined state play different roles.

Paper	Resources	Input modalities	Time	Venue
EgoSIS: From Factorized Visual Ego-Transitions to Motion-Canonical Spatial Evidence for UAV Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Interpretable Temporal Video Reasoning with EventGraph and EventField	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Unfold The World: Factorize 4D Properties in Reinforcing Spatial Reasoning	`N/A`	![Text][mod-text]	2026-09	`arXiv`
StateTrace: An Object-Centric Framework for Hidden-State Spatiotemporal Reasoning in Long Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Beyond Visual CoT: Internalized Visual Thinking for Proactive Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
SCOUT: Unlocking Enhanced Spatial Reasoning via Structured Chain-of-Thought and Multi-Objective Process Reward	`N/A`	![Text][mod-text]	2026-08	`arXiv`
R4DSG: Relative 4D Scene Graph Memory for Object-Centric Question Answering in Long Egocentric Video	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Multi-View Relational Distillation for Spatial Reasoning with Vision-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Listen, See and Track: Spatio-Temporal Audio-Visual Sound Event Reasoning for Omni-Modal Language Models	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-08	`arXiv`
Keep It Simple: Multi-Key Episodic Memory Retrieval for Ultra-Long Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
ChronoVision: Temporal Reasoning via Latent State Reconstruction	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
LAVE: Latent Visual Evidence-Enhanced Planning for Video Tool-use Agents	`N/A`	![Video][mod-video]	2026-08	`arXiv`
Perception Before Reasoning: Dynamic Latent Reasoning for Video Understanding and Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Evidence-Grounded Multimodal Knowledge Graph Construction for Multi-Lecture Educational Reasoning	`N/A`	![Text][mod-text]	2026-08	`arXiv`
Beyond Frame Selection: Generative Latent Evidence Aggregation for Long-Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
HyperClaim: Fine-Grained Cross-Modal Hypergraph Reasoning for Video Misinformation Detection	`N/A`	`N/A`	2026-07	`arXiv`
ViSAGE: Constructing Self-Correcting Memories for Long-Form Video Understanding	`N/A`	![Video][mod-video]	2026-07	`arXiv`
ChronoStitch: Training-Free Composition of Visual KV Memories for Long-Horizon Temporal Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
MEMORA: Embodied Action Memory from Egocentric Videos for Reasoning and Planning	![resource:code][res-github-code]	![Video][mod-video]	2026-07	`arXiv`
Whareformer: Learning to Track What is Where in Long Egocentric Videos	![resource:project][res-project]	![Video][mod-video]	2026-07	`arXiv`
Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
SpaceEra++: A Unified Framework Towards 3D Spatial Reasoning in Video	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Latent Visual Cache for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Reasoning with Memory: A Temporal Granularity-Adaptive Framework for Training-Free Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
VisReflect: Latent Visual Reflection for Fine-Grained Perception in Long Visual Context	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
DELTAVID: Enhancing Fine-Grained Spatiotemporal Perception with Cross-Video Differences	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Graph it first! Enabling Reasoning on Long-form Egocentric Videos through Scene Graphs	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
VideoLatent: Video-Language Learning via Latent Self-Forcing	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
ProPlay: Procedural World Models for Self-Evolving LLM Agents	![resource:code][res-github-code]	![Text][mod-text]	2026-06	`arXiv`
From Content to Knowledge: Lightning Fast Long-Video Understanding with Neural Knowledge Representations	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Plan-and-Verify Video Reward Reasoning with Spatio-Temporal Scene Graph Grounding	`N/A`	![Text][mod-text]	2026-06	`arXiv`
Reason, Then Re-reason: Cross-view Revisiting Improves Spatial Reasoning	![resource:project][res-project]	![Text][mod-text]	2026-06	`arXiv`
MemDreamer: Decoupling Perception and Reasoning for Long Video Understanding via Hierarchical Graph Memory and Agentic Retrieval Mechanism	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
MemoryCard: Topic-Aware Multi-Modal Clue Compression for Long-Video Question Answering	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Learning Geometric Representations from Videos for Spatial Intelligent Multimodal Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Imagine Before You Predict: Interleaved Latent Visual Reasoning for Video Event Prediction	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
GOPAgen: Motion-Aware and Efficient Agentic Long-Video Understanding with Structural Memory and Hierarchical Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Reasmory: 3D Reconstruction as Explicit Memory for VLMs Spatial Reasoning	`N/A`	![Text][mod-text]	2026-05	`arXiv`
Q-GeoMem: Question-Guided Geometric Memory for Video Spatial Reasoning	`N/A`	![Text][mod-text]	2026-05	`arXiv`
STORM: Internalized Modeling for Spatial-Temporal Reasoning in Video-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Cambrian-P: Pose-Grounded Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Bernini: Latent Semantic Planning for Video Diffusion	![resource:project][res-project]	![Text][mod-text]	2026-05	`arXiv`
LatentOmni: Rethinking Omni-Modal Understanding via Unified Audio-Visual Latent Reasoning	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-05	`arXiv`
Δynamics: Language-Based Representation for Inferring Rigid-Body Dynamics From Videos	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
LMM-Track4D: Eliciting 4D Dynamic Reasoning in LMMs via Trajectory-Grounded Dialogue	![resource:code][res-github-code]	![Text][mod-text]	2026-05	`arXiv`
PyraVid: Hierarchical Multimodal Memory for Long-Horizon Video Reasoning	`N/A`	![Video][mod-video]	2026-05	`arXiv`
LASAR: Towards Spatio-temporal Reasoning with Latent Cognitive Map	`N/A`	`N/A`	2026-05	`CVPR 2026`
SceneGraphVLM: Dynamic Scene Graph Generation from Video with Vision-Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
SpaceMind++: Toward Allocentric Cognitive Maps for Spatially Grounded Video MLLMs	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Tracking the Truth: Object-Centric Spatio-Temporal Monitoring for Video Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Bridging Modalities, Spanning Time: Structured Memory for Ultra-Long Agentic Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Event-Causal RAG: A Retrieval-Augmented Generation Framework for Long Video Reasoning in Complex Scenarios	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
4DThinker: Thinking with 4D Imagery for Dynamic Spatial Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Rethinking Temporal Consistency in Video Object-Centric Learning: From Prediction to Correspondence	`N/A`	`N/A`	2026-05	`arXiv`
Scaling Video Understanding via Compact Latent Multi-Agent Collaboration	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
ONOTE: Hypergraph-Grounded Omnimodal Reasoning for Computational Music Science	`N/A`	`N/A`	2026-04	`arXiv`
Frequency-guided Multi-level Reasoning for Scene Graph Generation in Video	`N/A`	`N/A`	2026-04	`arXiv`
VideoStir: Understanding Long Videos via Spatio-Temporally Structured and Intent-Aware RAG	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-04	`ACL 2026`
Graph-to-Frame RAG: Visual-Space Knowledge Fusion for Training-Free and Auditable Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`CVPR 2026`
A 4D Representation for Training-Free Agentic Reasoning from Monocular Laparoscopic Video	![resource:project][res-project]	![Text][mod-text]	2026-04	`arXiv`
Learning Multi-View Spatial Reasoning from Cross-View Relations	`N/A`	![Text][mod-text]	2026-03	`CVPR 2026`
SpatialStack: Layered Geometry-Language Fusion for 3D VLM Spatial Reasoning	![resource:project][res-project]	![Text][mod-text]	2026-03	`CVPR 2026`
Make Geometry Matter for Spatial Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Unleashing Spatial Reasoning in Multimodal Large Language Models via Textual Representation Guided Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`ACL 2026`
Reconstruction-Guided Slot Curriculum: Addressing Object Over-Fragmentation in Video Object-Centric Learning	`N/A`	`N/A`	2026-03	`CVPR 2026`
Motion-o: Trajectory-Grounded Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Clue Matters: Leveraging Latent Visual Clues to Empower Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Towards Stable Self-Supervised Object Representations in Unconstrained Egocentric Video	`N/A`	![Video][mod-video]	2026-03	`CVPR 2026`
Towards Spatio-Temporal World Scene Graph Generation from Monocular Videos	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
EgoReasoner: Learning Egocentric 4D Reasoning via Task-Adaptive Structured Thinking	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Thinking with Spatial Code for Physical-World Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
FocusGraph: Graph-Structured Frame Selection for Embodied Long Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
ViterbiPlanNet: Injecting Procedural Knowledge via Differentiable Viterbi for Planning in Instructional Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
SCP: Spatial Causal Prediction in Video	![resource:project][res-project]	`N/A`	2026-03	`arXiv`
EgoGraph: Temporal Knowledge Graph for Egocentric Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
TrajTok: Learning Trajectory Tokens enables better Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Keep it SymPL: Symbolic Projective Layout for Allocentric Spatial Reasoning in Vision-Language Models	`N/A`	![Text][mod-text]	2026-02	`CVPR 2026`
GraphThinker: Reinforcing Temporally Grounded Video Reasoning with Event Graph Thinking	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
AD-MIR: Bridging the Gap from Perception to Persuasion in Advertising Video Understanding via Structured Reasoning	![resource:code][res-github-code]	![Video][mod-video]	2026-02	`arXiv`
MosaicThinker: On-Device Visual Spatial Reasoning for Embodied AI via Iterative Construction of Space Representation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Spatial-Conditioned Reasoning in Long-Egocentric Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
SpatialMem: Metric-Aligned Long-Horizon Video Memory for Language Grounding and QA	`N/A`	![Video][mod-video]	2026-01	`arXiv`
MMViR: A Multi-Modal and Multi-Granularity Representation for Long-range Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
AbductiveMLLM: Boosting Visual Abductive Reasoning Within MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Semantic Event Graphs for Long-Form Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Learning to Reason in 4D: Dynamic Spatial Understanding for Vision Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`
4D-RGPT: Toward Region-level 4D Understanding via Perceptual Distillation	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`

💻 Executable Worlds

Executable state, transitions, simulators, and learned rendering. Includes direct video systems and adjacent visual/planning foundations; see the case study.

Paper	Resources	Input modalities	Time	Venue
Recursive Code World Models: Building Complex Worlds through Recursive Scene Programs	`N/A`	![Image][mod-image]	2026-09	`arXiv`
Programmable World Model	![resource:project][res-project] ![resource:code][res-github-code]	![Text][mod-text]	2026-09	`arXiv`
Code as Worlds: Agentic Discovery of Executable World Representations for Physical Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Code World Model: Coding Agent as World Brain	![resource:project][res-project] ![resource:code][res-github-code] ![resource:data][res-hf-data] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Image][mod-image]	2026-08	`arXiv`
Compiling VGDL into Causal Models	`N/A`	![Text][mod-text]	2026-08	`arXiv`
AppDeltaWorld: Transition-Grounded Delta Code World Model for Mobile GUI Agents	`N/A`	`N/A`	2026-08	`arXiv`
PhysMind: From Video to Executable Worlds for Training-Free Physical Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Tycho: Active Abstraction with Programmatic World Models for ARC-AGI-3	![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
VideoCoCo: Code-as-CoT for Physically-Consistent Video Generation via an Agentic Dual-Engine System	`N/A`	![Text][mod-text]	2026-07	`arXiv`
VisualPatchWorld: Code World Models as Latent Structured Representations for Planning	![resource:code][res-github-code]	`N/A`	2026-07	`arXiv`
Agentic Real2Sim: Physics-based World Modeling with Vision-Language Agents	![resource:project][res-project]	![Text][mod-text]	2026-07	`arXiv`
PhysAgent: Reflective Agentic Physics Control for Physically Plausible Video Generation	![resource:project][res-project]	![Text][mod-text]	2026-07	`arXiv`
Do Coding Agents Need Executable World Models, Simplification, and Verification to Solve ARC-AGI-3?	`N/A`	![Text][mod-text]	2026-07	`arXiv`
When a Verified World Model Still Loses: Play-Adequacy vs Prediction-Accuracy in LLM-Synthesized Code World Models	![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
World-Time Compute with Verified Code World Models	![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration for ARC-AGI-3	`N/A`	![Text][mod-text]	2026-07	`arXiv`
World Narrative Model for Highly Controllable Video Generation: A Paradigm Shift from Pixel Sampling to Physical World Orchestration	`N/A`	`N/A`	2026-06	`arXiv`
STRIPS-WM: Learning Grounded Propositional STRIPS-style World Models from Images	`N/A`	![Image][mod-image]	2026-06	`arXiv`
PatchWorld: Gradient-Free Optimization of Executable World Models for Agent Environments	![resource:code][res-github-code]	![Text][mod-text]	2026-05	`arXiv`
Distilling Game Code World Model Generation into Lightweight Large Language Models	`N/A`	![Text][mod-text]	2026-05	`arXiv`
SWoMo: Neuro-Symbolic World Model for Cataract Surgery Simulation	![resource:project][res-project]	`N/A`	2026-05	`arXiv`
Ego2World: Compiling Egocentric Cooking Videos into Executable Worlds for Belief-State Planning	`N/A`	![Video][mod-video]	2026-05	`arXiv`
Executable World Models for ARC-AGI-3 in the Era of Coding Agents	![resource:code][res-github-code]	![Text][mod-text]	2026-05	`arXiv`
Physical Simulator In-the-Loop Video Generation	`N/A`	`N/A`	2026-03	`CVPR 2026`
SimRecon: SimReady Compositional Scene Reconstruction from Real Videos	![resource:project][res-project]	![Video][mod-video]	2026-03	`CVPR 2026`
OOWM: Structuring Embodied Reasoning and Planning via Object-Oriented Programmatic World Modeling	`N/A`	![Text][mod-text]	2026-02	`arXiv`
Neuro-Symbolic Synergy for World Modeling	![resource:code][res-github-code]	![Text][mod-text]	2026-02	`arXiv`
Code2World: A GUI World Model via Renderable Code Generation	![resource:code][res-github-code] ![resource:project][res-project]	![Text][mod-text] ![Action][mod-action]	2026-02	`arXiv`
Generative Visual Code Mobile World Models	`N/A`	![Text][mod-text]	2026-02	`arXiv`
CASSANDRA: Programmatic and Probabilistic Learning and Inference for Stochastic World Modeling	`N/A`	![Text][mod-text]	2026-01	`arXiv`
Agent2World: Learning to Generate Symbolic World Models via Adaptive Multi-Agent Feedback	![resource:project][res-project]	![Text][mod-text]	2025-12	`arXiv`
Generating Code World Models with Large Language Models Guided by Monte Carlo Tree Search	`N/A`	![Text][mod-text]	2024-05	`arXiv`
WorldCoder, a Model-Based LLM Agent: Building World Models by Writing Code and Interacting with the Environment	`N/A`	![Text][mod-text]	2024-02	`arXiv`

🔄 Interaction Loops

🛠️ Active Video Agents

Tool use, evidence search, reflection, and agent coordination. Full-recording retrieval remains an offline access setting.

Paper	Resources	Input modalities	Time	Venue
VideoScout: Learning Agentic Active Exploration with Adaptive Reasoning Pacing for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Online Video Agent Harness for Long Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
One Skill Does Not Fit All: Automatic Discovery and Taxonomy-Guided Routing of Frame-Selection Skills for Long-Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Caption-once, Frames-on-Demand: Visual-Need Routing for Budget-Aware Agentic Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
VLX-VR: An Agentic-Aware Video Reasoning Model	`N/A`	![Video][mod-video]	2026-09	`arXiv`
Thinking with Cameras: Active Visual Reasoning via Dynamic Viewpoint Control for Surveillance Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Visual Search Augmented Chain-of-Thought Reasoning for Attribute Value Extraction from Product Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
From Intent to Evidence: Policy-Steered Multi-Strategy Retrieval for Long-Video Agents	`N/A`	`N/A`	2026-08	`arXiv`
Temporal Tree of Thought: Reasoning-Guided Visual Cue Search for Long-Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
AdaVDR: Adaptive Tool Use and Reflection for Video Deep Research	`N/A`	![Video][mod-video]	2026-08	`arXiv`
VideoHarness-RSI: Recursive Harness Self-Improvement for Long-Video Understanding with Frozen Vision-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Thinking Beyond Videos: Unifying Video Reasoning and Deep Research for Open-World Video Agents	`N/A`	![Video][mod-video]	2026-08	`arXiv`
Coarse Indexing, Fine Evidence: Decoupling Temporal Granularity in Long-Video RAG	`N/A`	![Text][mod-text]	2026-08	`arXiv`
TRACE: Temporal Retrieval with Anchored and Convergent Evidence for Long-Horizon Video Understanding	![resource:project][res-project]	![Video][mod-video]	2026-08	`arXiv`
Routing Before Looking: Query-Adaptive Evidence Acquisition for Long-form Video Understanding	`N/A`	![Video][mod-video]	2026-08	`arXiv`
MedClaw: Heuristic Agent Harness for Long-Horizon Surgical Video Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
REVEAL: A Rubric-Guided Agent for Explicit Evidence Sufficiency Verificationin Long-Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
VTO: Visual Tool Orchestration for Video Anomaly Detection	![resource:code][res-github-code]	![Video][mod-video]	2026-08	`arXiv`
SCOUT: Self-Checking and Recovery-Aware Tool-Thought Agents for Ultra-Long Egocentric Video Reasoning	`N/A`	![Video][mod-video]	2026-08	`arXiv`
SportsGrounder: Proposal-Aided Interleaved Grounding for Dense Sports Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Glance, Scrutinize, and Think: Advancing Video Anomaly Detection from Training-Free to Agentic Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
DAEP: Difficulty-Aware Evidence Planning for Medical Video Corpus Temporal Answer Grounding	`N/A`	`N/A`	2026-08	`arXiv`
Multi-Agent Forensic Reasoning for Generalizable Deepfake Video Detection	![resource:project][res-project]	![Text][mod-text]	2026-08	`arXiv`
MetaVideoAgent: Automated Video-Agent Evolution for Long-Form Video Understanding	![resource:code][res-github-code]	![Video][mod-video]	2026-08	`arXiv`
Video-DeepResearch: Towards the Next-Generation Multimodal Deepresearch Agent	![resource:code][res-github-code]	![Text][mod-text]	2026-08	`arXiv`
When and Where to Look: Adaptive Visual Evidence Scheduling for Efficient Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
AgenticVAU: Multi-Agent Explore-Verify Reasoning for Video Anomaly Understanding	`N/A`	![Video][mod-video]	2026-08	`arXiv`
ReflexTrack: A Feedback-Driven Agent for Training-Free Referring Video Object Segmentation	`N/A`	![Text][mod-text]	2026-07	`arXiv`
MarineEVT: Advancing Event-Centric Marine Video Understanding via Visual Tool Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
OmniReasoner: Thinking with Long Audio-Video via Native Tool Use	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2026-07	`arXiv`
Sparse Evidence Can Suffice: Agentic Evidence Seeking for Multimodal Video Misinformation Detection	`N/A`	`N/A`	2026-07	`arXiv`
Searching Videos as Trees: Self-Correcting Agents for Grounded Long Video QA	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Beyond the Single Camera: Agentic Multi-View Reasoning in Sports Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
TreeSoc: Tree-Structured Dynamic Reasoning and Tool Synergy for Soccer Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
SafeGuard: A Multi-Agent Perception-Reasoning Framework for Social-Risk AI-Generated Video Detection	![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
Incentivizing Vision Language Models to Search for Long Video Question Answering	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
VideoSearcher: Empowering Video Deep Research with Multi-Tool Agentic Reasoning via Reinforcement Learning	![resource:project][res-project] ![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
Learning to Watch: Active Video Anomaly Understanding via Interleaved Policy Optimization	`N/A`	![Video][mod-video]	2026-07	`arXiv`
Homer: Understanding Long-form Videos with Hierarchical Memory and Agentic Reasoning	`N/A`	![Text][mod-text]	2026-07	`arXiv`
VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Confidence-Aware Tool Orchestration for Robust Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
video-SALMONN-R³: Learning to ReWatch, ReAsk, and ReAnswer for Efficient Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
VideoAgent: All-in-One Framework for Video Understanding and Editing	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
HPP: Hierarchical Programmatic Probing for Long Video Understanding by Decoupling Perception and Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
TimeProVe: Propose, then Verify for Efficient Long Video Temporal Reasoning in Activities of Daily Living	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence	![resource:project][res-project]	![Text][mod-text]	2026-06	`arXiv`
Native Active Perception as Reasoning for Omni-Modal Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning	`N/A`	`N/A`	2026-06	`arXiv`
Orchestra-o1: Omnimodal Agent Orchestration	`N/A`	![Text][mod-text]	2026-06	`arXiv`
MAVIS: Multi-Agent Video Retrieval via Structured Video Understanding	`N/A`	![Video][mod-video]	2026-06	`ACL 2026 (Findings)`
VTI-CoT: Visual-Textual Interleaved Chain of Thought for Video Reasoning	`N/A`	![Video][mod-video]	2026-06	`arXiv`
AgentCVR: Active Multi-Agent Cross-Video Reasoning via Script-Simulated Reinforcement Learning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Agentic Active Omni-Modal Perception for Multi-Hop Audio-Visual Reasoning	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-05	`arXiv`
DynFrame: Adaptive Reasoning-Driven Multimodal Framework with Dynamic Frame Augmentation for Complex Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Decomposing Queries into Tool Calls for Long-Video Keyframe Retrieval	![resource:code][res-github-code]	![Text][mod-text]	2026-05	`arXiv`
ParaVT: Taming the Tool Prior Paradox for Parallel Tool Use in Agentic Video Reinforcement Learning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Seeing Together: Multi-Robot Cooperative Egocentric Spatial Reasoning with Multimodal Large Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
VideoSeeker: Incentivizing Instance-level Video Understanding via Native Agentic Tool Invocation	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
EvoGround: Self-Evolving Video Agents for Video Temporal Grounding	![resource:project][res-project]	![Video][mod-video]	2026-05	`arXiv`
ReTool-Video: Recursive Tool-Using Video Agents with Meta-Augmented Tool Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
VideoSEAL: Mitigating Evidence Misalignment in Agentic Long Video Understanding by Decoupling Answer Authority	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
ViSRA: A Video-based Spatial Reasoning Agent for Multi-modal Large Language Models	`N/A`	![Text][mod-text]	2026-05	`arXiv`
CollabVR: Collaborative Video Reasoning with Vision-Language and Video Generation Models	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
AffectSeek: Agentic Affective Understanding in Long Videos under Vague User Queries	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
VTAgent: Agentic Keyframe Anchoring for Evidence-Aware Video TextVQA	`N/A`	![Text][mod-text]	2026-05	`arXiv`
Video Active Perception: Effective Inference-Time Long-Form Video Understanding with Vision-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Act2See: Emergent Active Visual Perception for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`CVPR 2026`
Omni-o3: Deep Nested Omnimodal Deduction for Deliberative Audio-Visual Reasoning	`N/A`	![Audio][mod-audio]	2026-04	`arXiv`
Zoom In, Reason Out: Efficient Far-field Anomaly Detection in Expressway Surveillance Videos via Focused VLM Reasoning Guided by Bayesian Inference	`N/A`	![Text][mod-text]	2026-04	`arXiv`
HiCrew: Hierarchical Reasoning for Long-Form Video Understanding via Question-Aware Multi-Agent Collaboration	`N/A`	![Video][mod-video]	2026-04	`arXiv`
Video-ToC: Video Tree-of-Cue Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
IMPACT-CYCLE: A Contract-Based Multi-Agent System for Claim-Level Supervisory Correction of Long-Video Semantic Memory	![resource:code][res-github-code]	![Video][mod-video]	2026-04	`arXiv`
Chain-of-Glimpse: Search-Guided Progressive Object-Grounded Reasoning for Video Understanding	`N/A`	![Video][mod-video]	2026-04	`arXiv`
Open-Ended Video Game Glitch Detection with Agentic Reasoning and Temporal Grounding	`N/A`	`N/A`	2026-04	`arXiv`
RASR: Retrieval-Augmented Semantic Reasoning for Fake News Video Detection	`N/A`	![Text][mod-text]	2026-04	`arXiv`
SVAgent: Storyline-Guided Long Video Understanding via Cross-Modal Multi-Agent Collaboration	`N/A`	![Video][mod-video]	2026-04	`CVPR 2026`
QVAD: A Question-Centric Agentic Framework for Efficient and Training-Free Video Anomaly Detection	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Progressive Video Condensation with MLLM Agent for Long-form Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
MAR3: Multi-Agent Recognition, Reasoning, and Reflection for Reference Audio-Visual Segmentation	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-03	`arXiv`
VideoTIR: Accurate Understanding for Long Videos with Efficient Tool-Integrated Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
LensWalk: Agentic Video Understanding by Planning How You See in Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
AgentRVOS: Reasoning over Object Tracks for Zero-Shot Referring Video Object Segmentation	![resource:project][res-project]	![Text][mod-text]	2026-03	`arXiv`
EVA: Efficient Reinforcement Learning for End-to-End Video Agent	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
VideoDetective: Clue Hunting via both Extrinsic Query and Intrinsic Relevance for Long Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
VideoSeek: Long-Horizon Video Agent with Tool-Guided Seeking	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
VideoAtlas: Navigating Long-Form Video in Logarithmic Compute	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Interpretable Traffic Responsibility from Dashcam Video via Legal Multi Agent Reasoning	`N/A`	![Text][mod-text]	2026-03	`arXiv`
Symphony: A Cognitively-Inspired Multi-Agent System for Long-Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
A Skill-augmented Agentic Framework and Benchmark for Multi-Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
A Multi-Agent Perception-Action Alliance for Efficient Long Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
Think, Then Verify: A Hypothesis-Verification Multi-Agent Framework for Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
LLandMark: A Multi-Agent Framework for Landmark-Aware Multimodal Interactive Video Retrieval	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Training-Free Spatio-temporal Decoupled Reasoning Video Segmentation with Adaptive Object Memory	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
FactGuard: Agentic Video Misinformation Detection via Reinforcement Learning	`N/A`	![Text][mod-text]	2026-02	`arXiv`
LongVideo-R1: Smart Navigation for Low-cost Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-02	`CVPR 2026`
VIGiA: Instructional Video Guidance via Dialogue Reasoning and Retrieval	`N/A`	![Video][mod-video]	2026-02	`arXiv`
Towards Sparse Video Understanding and Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-02	`CVPR 2026`
MedScope: Incentivizing "Think with Videos" for Clinical Reasoning via Coarse-to-Fine Tool Calling	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Agentic Spatio-Temporal Grounding via Collaborative Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
VideoTemp-o3: Harmonizing Temporal Grounding and Video Understanding in Agentic Thinking-with-Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Weaver: End-to-End Agentic System Training for Video Interleaved Reasoning	`N/A`	![Video][mod-video]	2026-02	`arXiv`
VideoBrain: Learning Adaptive Frame Sampling for Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
OmniRAG-Agent: Agentic Omnimodal Reasoning for Low-Resource Long Audio-Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2026-02	`arXiv`
Refer-Agent: A Collaborative Multi-Agent System with Reasoning and Reflection for Referring Video Object Segmentation	![resource:code][res-github-code]	![Text][mod-text]	2026-02	`CVPR 2026`
Video-o3: Native Interleaved Clue Seeking for Long Video Multi-Hop Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Agentic Very Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-01	`ACL 2026`
VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Hierarchical Long Video Understanding with Audiovisual Entity Cohesion and Agentic Search	`N/A`	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2026-01	`CVPR 2026`
TIR-Flow: Active Video Search and Reasoning with Frozen VLMs	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
FastV-RAG: Towards Fast and Fine-Grained Video QA with Retrieval-Augmented Generation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`ACL 2026`
Active Perception Agent for Omnimodal Audio-Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-12	`arXiv`
Video-Browser: Towards Agentic Open-web Video Browsing	![resource:code][res-github-code]	![Video][mod-video]	2025-12	`arXiv`
VideoZoomer: Reinforcement-Learned Temporal Focusing for Long Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
LongVideoAgent: Multi-Agent Reasoning with Long Videos	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2025-12	`ACL 2026`
Video Detective: Seek Critical Clues Recurrently to Answer Question from Long Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
AdaTooler-V: Adaptive Tool-Use for Images and Videos	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Image][mod-image]	2025-12	`ACL 2026 (Findings)`
LongVT: Incentivizing "Thinking with Long Videos" via Native Tool Calling	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-11	`CVPR 2026`
Video-R4: Reinforcing Text-Rich Video Reasoning with Visual Rumination	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
Orion: A Unified Visual Agent for Multimodal Perception, Advanced Visual Reasoning and Execution	`N/A`	![Text][mod-text]	2025-11	`arXiv`
Video-in-the-Loop: Span-Grounded Long Video QA with Interleaved Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-10	`arXiv`
FrameMind: Frame-Interleaved Video Reasoning via Reinforcement Learning	`N/A`	![Video][mod-video]	2025-09	`arXiv`
Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video Reasoning	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-08	`arXiv`
ViTCoT: Video-Text Interleaved Chain-of-Thought for Boosting Video Understanding in Large Language Models	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-07	`ACM-MM 2025`

🔁 Unified Understanding & Generation

Architectures combining understanding and generation; iterative reasoning requires separate evidence.

Paper	Resources	Input modalities	Time	Venue
JavisGPT: A Unified Multi-modal LLM for Sounding-Video Comprehension and Generation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-11	`NeurIPS 2025 (Spotlight)`
VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generation	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text]	2024-09	`ICLR 2025`

📡 Always-On Intelligence

Observed-prefix reasoning, persistent memory, response timing, and proactive interaction.

Paper	Resources	Input modalities	Time	Venue
SVMemAgent: A Streaming Video Memory Agent for Query-Agnostic Online Frame Selection	`N/A`	![Video][mod-video]	2026-09	`arXiv`
Omni-Streaming Thinking	`N/A`	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2026-09	`arXiv`
ReactVAU: A Slow-Fast Decoupled Framework for Streaming Video Anomaly Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Beyond Retrieval: Progressive Latent Memory Evolution for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
CoFiE: Coarse-to-Fine Evidence Selection for Efficient Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
ShallowStream: Index Shallow then Answer Deep for Streaming Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
StreamScout: Learning When to Look Deeper for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Dynamic Hub-and-Spoke Memory for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
StreamEMS: Streaming Video Understanding with Self-Evolving Memory Scheme for Vision-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Emotion Understanding in Streaming Video with Trajectory-Aware Reliability	`N/A`	![Video][mod-video]	2026-08	`arXiv`
Strictly Causal Streaming Video Anomaly Detection with a Theoretically-Grounded State-Space Core	`N/A`	![Video][mod-video]	2026-08	`arXiv`
StreamOPD: A Post-Training Recipe with Spatio-Temporal Cue Gating for Streaming Video Understanding	![resource:project][res-project]	![Video][mod-video]	2026-08	`arXiv`
StreamFlow: Dynamic Memory Flows for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
GROVE: Growing and Reasoning over Temporally Stratified Memory from Streaming Video Experience	![resource:code][res-github-code]	![Video][mod-video]	2026-08	`arXiv`
Think in Sets for Streaming Video Token Compression	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
ObjectStream: Latent Objects as Memory Anchors for Streaming Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Memory-Augmented Multimodal Large Language Models for Small Object Understanding in Streaming Aerial Videos	`N/A`	![Text][mod-text]	2026-07	`arXiv`
FOLIO: Focused Semantic Memory for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Vinci2: Providing Proactive Assistance in Continuous Egocentric Videos	![resource:project][res-project]	![Video][mod-video]	2026-07	`arXiv`
Light-Omni: Reflex over Reasoning in Agentic Video Understanding with Long-Term Memory	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
ReflectWorld-MM: An Entity-Oriented Multimodal Memory System for Open-Ended Video Streams	`N/A`	![Video][mod-video]	2026-07	`arXiv`
GuideMe: Multi-Domain Task Guidance and Intervention in Streaming Video	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
ProtoKV: Streaming Video Understanding under Delayed Query with Summary-State Memory	`N/A`	![Video][mod-video]	2026-06	`arXiv`
Towards a Dynamic and Fixed-budget Memory Bank for Efficient Streaming Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
ESTANet: Efficient Online Error Detection in Procedural Videos via Prediction Inconsistency	`N/A`	`N/A`	2026-06	`arXiv`
CoVStream: Edge-Cloud Collaboration for Understanding of Long Video Streams	`N/A`	![Video][mod-video]	2026-06	`arXiv`
LiveStarPro: Proactive Streaming Video Understanding with Hierarchical Memory for Long-Horizon Streams	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
What Should a Streaming Video Model Remember?	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Keep It in Mind: User Centric Continual Spatial Intelligence Reasoning in Egocentric Video Streams	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
AdaSR: Adaptive Streaming Reasoning with Hierarchical Relative Policy Optimization	![resource:code][res-github-code]	![Text][mod-text] ![Audio][mod-audio]	2026-06	`arXiv`
DuplexOmni: Real-Time Listening, Seeing, Thinking, and Speaking for Full-Duplex Interaction	`N/A`	![Video][mod-video] ![Audio][mod-audio]	2026-06	`arXiv`
Harnessing Streaming Video in the Wild	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Don't Pause: Streaming Video-Language Synchrony for Online Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
MOSS-Video-Preview: Toward Real-Time Video Understanding via Cross-Attention	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
FlowNar: Scalable Streaming Narration for Long-Form Videos	![resource:code][res-github-code]	![Text][mod-text]	2026-05	`arXiv`
OmniMem: Perturbation-aware Memory Compression for Streaming Audio-Visual LLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2026-05	`arXiv`
StreamOV: Streaming Omni-Video Understanding via Evidence-Guided Memory and Response Triggering	`N/A`	![Video][mod-video] ![Audio][mod-audio]	2026-05	`arXiv`
MuKV: Multi-Grained KV Cache Compression for Long Streaming Video Question-Answering	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`CVPR 2026`
An Efficient Streaming Video Understanding Framework with Agentic Control	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
CoRDS: Coreset-based Representative and Diverse Selection for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
EvoStreaming: Your Offline Video Model Is a Natively Streaming Assistant	`N/A`	![Text][mod-text]	2026-05	`arXiv`
StreamPro: From Reactive Perception to Proactive Decision-Making in Streaming Video	`N/A`	![Video][mod-video]	2026-05	`arXiv`
Semantic-Aware Adaptive Visual Memory for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Response-G1: Explicit Scene Graph Modeling for Proactive Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`ACL 2026`
Decouple and Cache: KV Cache Construction for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Progressive Online Video Understanding with Evidence-Aligned Timing and Transparent Decisions	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
OASIS: On-Demand Hierarchical Event Memory for Streaming Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-04	`CVPR 2026`
Online Reasoning Video Object Segmentation	`N/A`	`N/A`	2026-04	`arXiv`
StreamMeCo: Long-Term Agent Memory Compression for Efficient Streaming Video Understanding	![resource:code][res-github-code]	![Video][mod-video]	2026-04	`ACL 2026 (Findings)`
ESOM: Efficiently Understanding Streaming Video Anomalies with Open-world Dynamic Definitions	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference	`N/A`	![Text][mod-text]	2026-04	`arXiv`
AURA: Always-On Understanding and Real-Time Assistance via Video Streams	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
A Simple Baseline for Streaming Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
STRIDE: When to Speak Meets Sequence Denoising for Streaming Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
GazeQwen: Lightweight Gaze-Conditioned LLM Modulation for Streaming Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
StreamingClaw Technical Report	`N/A`	![Video][mod-video]	2026-03	`arXiv`
PEARL: Personalized Streaming Video Understanding Model	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
CurveStream: Boosting Streaming Video Understanding in MLLMs via Curvature-Aware Hierarchical Visual Memory Management	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Em-Garde: A Propose-Match Framework for Proactive Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Thinking in Streaming Video	![resource:code][res-github-code]	![Video][mod-video]	2026-03	`arXiv`
Video Streaming Thinking: VideoLLMs Can Watch and Think Simultaneously	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`ECCV 2026`
WAT: Online Video Understanding Needs Watching Before Thinking	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Think While Watching: Online Streaming Segment-Level Memory for Multi-Turn Video Reasoning in Multimodal Large Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
StreamReady: Learning What to Answer and When in Long Streaming Videos	`N/A`	![Video][mod-video]	2026-03	`CVPR 2026`
Proact-VL: A Proactive VideoLLM for Real-Time AI Companions	`N/A`	![Text][mod-text]	2026-03	`arXiv`
Think-as-You-See: Streaming Chain-of-Thought Reasoning for Large Vision-Language Models	![resource:code][res-github-code]	![Text][mod-text]	2026-03	`CVPR 2026`
FluxMem: Adaptive Hierarchical Memory for Streaming Video Understanding	`N/A`	![Video][mod-video]	2026-03	`CVPR 2026`
From Verbatim to Gist: Distilling Pyramidal Multimodal Memory via Semantic Information Bottleneck for Long-Horizon Video Agents	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`ACL 2026`
RT-NeuS: Towards Real-Time Neuro-Symbolic Video Understanding via Adaptive Temporal Verification	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
WeaveTime: Stream from Earlier Frames into Emergent Memory in VideoLLMs	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-02	`CVPR 2026`
Going Down Memory Lane: Scaling Tokens for Video Stream Understanding with Dynamic KV-Cache Memory	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
EventMemAgent: Hierarchical Event-Centric Memory for Online Video Understanding with Adaptive Tool Use	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Vista: Scene-Aware Optimization for Streaming Video Question Answering under Post-Hoc Queries	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
FreshMem: Brain-Inspired Frequency-Space Hybrid Memory for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Event-VStream: Event-Driven Real-Time Understanding for Long Video Streams	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
HERMES: KV Cache as Hierarchical Memory for Efficient Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`ACL 2026`
ROMA: Real-time Omni-Multimodal Assistant with Interactive Streaming Understanding	![resource:project][res-project]	![Text][mod-text]	2026-01	`ACL 2026 (Findings)`
Streaming Video Instruction Tuning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`
VideoScaffold: Elastic-Scale Visual Hierarchies for Streaming Video Understanding in MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
LiveStar: Live Streaming Assistant for Real-World Online Video Understanding	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-11	`NeurIPS 2025`
StreamingCoT: A Dataset for Temporal Dynamics and Multimodal Chain-of-Thought Reasoning in Streaming VideoQA	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-10	`ACM-MM 2025`
StreamingVLM: Real-Time Understanding for Infinite Video Streams	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-10	`ICLR 2026`
StreamForest: Efficient Online Video Understanding with Persistent Event Memory	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-09	`NeurIPS 2025 (Spotlight)`
StreamAgent: Towards Anticipatory Agents for Streaming Video Understanding	`N/A`	![Video][mod-video]	2025-08	`arXiv`
StreamVLN: Streaming Vision-and-Language Navigation via SlowFast Context Modeling	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text]	2025-07	`ICRA 2026`
Flash-VStream: Efficient Real-Time Understanding for Long Video Streams	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-06	`ICCV 2025`
LiveVLM: Efficient Online Video Understanding via Streaming-Oriented KV Cache and Retrieval	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-05	`DAC 2026`
StreamBridge: Turning Your Offline Video Large Language Model into a Proactive Streaming Assistant	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`NeurIPS 2025`
TimeChat-Online: 80% Visual Tokens are Naturally Redundant in Streaming Videos	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-04	`ACM-MM 2025`
LiveCC: Learning Video LLM with Streaming Speech Transcription at Scale	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-04	`CVPR 2025`
ViSpeak: Visual Instruction Feedback in Streaming Videos	![resource:code][res-github-code] ![resource:weights][res-github-weights]	![Text][mod-text] ![Video][mod-video]	2025-03	`ICCV 2025`
StreamMind: Unlocking Full Frame Rate Streaming Video Dialogue through Event-Gated Cognition	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-03	`ICCV 2025`
Streaming Video Question-Answering with In-context Video KV-Cache Retrieval	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-03	`ICLR 2025`
SVBench: A Benchmark with Temporal Multi-Turn Dialogues for Streaming Video Understanding	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-02	`ICLR 2025 (Spotlight)`
Streaming Video Understanding and Multi-round Interaction with Memory-enhanced Knowledge	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-01	`ICLR 2025`
Dispider: Enabling Video LLMs with Active Real-Time Interaction via Disentangled Perception, Decision, and Reaction	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-01	`CVPR 2025`
Online Video Understanding: OVBench and VideoChat-Online	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2024-12	`CVPR 2025`
StreamChat: Chatting with Streaming Video	`N/A`	![Text][mod-text] ![Video][mod-video]	2024-12	`arXiv`

🔎 Evidence & Action

🔎 Evidence at Scale

Frame selection, token compression, temporal grounding, and audio-visual representations supporting reasoning.

Paper	Resources	Input modalities	Time	Venue
VideoMM: Adaptive Macro-Micro Inference for Efficient Video MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
MarKey: Marginal Utility Guided Greedy Keyframe Selection for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
IAE-VTG: Interaction-Aligned Action-Entity Video Temporal Grounding	`N/A`	![Video][mod-video]	2026-09	`arXiv`
DSE-VTG: Dual-Side Enhancement for Training-Free Video Temporal Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Select, Compress, Reinvest: A Controlled Study of Visual-Token Allocation in Long-Video MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
RIDGE: Region-Informed Derivative-Guided Evidence Selection for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Locate Anything in Videos: Rethinking Efficient Generative Spatio-Temporal Video Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Training-Free Temporal Abstraction for General Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
[LongVU-TTT: Causal Test-Time Training for Visual Resampling in Long Video Understanding](h

Truncated — view the full README on GitHub.

artificial-intelligence

awesome

benchmarks

chain-of-frames

chain-of-thought

large-language-models

multimodal-large-language-models

omni

paper-list

real-time-video-understanding

streaming-video-reasoning

streaming-video-understanding

think-with-video

video-generation

video-reasoning

world-model

Contributors

119 commits

18 commits

3 commits

2 commits

Similar repos

yunlong10/Awesome-Video-LMM-Post-Training

🔥🔥🔥 Latest Papers, Codes and Datasets on Video-LMM Post-Training

303

yeliudev/VideoMind

🧠 VideoMind: A Chain-of-LoRA Agent for Temporal-Grounded Video Reasoning (ICLR 2026)

357

The-Martyr/Awesome-Multimodal-Reasoning

Latest Advances on (RL based) Multimodal Reasoning and Generation in Multimodal LLMs

102

HITsz-TMG/Awesome-Large-Multimodal-Reasoning-Models

The development and future prospects of large multimodal reasoning models.

616

lwpyh/Awesome-MLLM-Reasoning-Collection

A collection of multimodal reasoning papers, codes, datasets, benchmarks and resources.

fnlp/VideoThinkBench

Dataset

[CVPR 2026] Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm

yaotingwangofficial/Awesome-MCoT

Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey

1,026

AIWorldLab/Awesome-Vision-World-Model

From Seeing to Knowing the World: A Survey of Vision World Models

193

Languages

Python

95.8%

JavaScript

3.6%

LJungang/Awesome-Video-Reasoning-Landscape

🔥An open-source survey of the latest video reasoning tasks, paradigms, and benchmarks.

Python

191

148 commits

updated Sep 24, 2026

See the code

README

Video Reasoning Landscape

🗺️ At a Glance

A curated survey of reasoning over videos, reasoning through visual generation, and building worlds for prediction and action.

Browse the Paper Atlas, choose a benchmark, or contribute a paper.

📖 Contents

🧭 Start Here
📚 Paper Atlas
🎯 Choose Your Benchmark
🔬 Evaluation Playbook
🔭 Next Frontiers
📐 Modeling Notes
✈️ Further Reading
🤝 Contributing
🌟 Star History
♥️ Contributors

🧭 Start Here: Choose Your Research Track

Your question	Read	Compare on
Can a model answer from the right video evidence?	Grounded Reasoning · Stateful Intelligence	CaST-Bench, SagaQA: answer correctness and evidence localization
Does generating a visual trajectory help solve a problem?	Thinking in Visual Futures	VBVR-Pro, MME-CoF-Pro: intermediate rule validity and final success
Can a model construct and execute a consistent world?	Executable Worlds	WorldCoder-Bench: runtime state, physical constraints, and interaction contracts
Can an assistant search, remember, and respond at the right time?	Active Video Agents · Always-On Intelligence	StreamArena, ProactiveBench: observed-prefix access, memory, and response timing
Do imagined futures improve action selection?	World Models for Action	WorldArena: rollout fidelity and downstream decision utility

📚 Paper Atlas

Time is the first-public month; (proc.) marks a proceedings date with an unresolved earlier preprint. N/A means unrecorded or not applicable.

Input modalities & benchmark tasks

Badges describe supplied inputs, including optional conditioning. They exclude outputs, internal representations, and task names. Lists cover verified inputs and may be incomplete.

Modality	Input	Color
![Text][mod-text]	Questions, instructions, system prompts, transcripts, or textual context	`#AEC6DF`
![Video][mod-video]	Observed clips, streams, or temporally ordered frames	`#C3E6CB`
![Audio][mod-audio]	Waveforms, speech, or environmental sound	`#F7D6AE`
![Image][mod-image]	Still images or reference images	`#D7C4E8`
![Depth][mod-depth]	Depth maps or RGB-D observations	`#F2C6D6`
![Point Cloud][mod-point-cloud]	Observed 3D point sets	`#B9E3E0`
![Action][mod-action]	Control commands or action sequences supplied to a model	`#F6E3A9`
![State][mod-state]	Supplied object, environment, or robot states and structured annotations	`#D8DEE9`
![Code][mod-code]	Source code or programs supplied as input	`#C9C9F1`
![Mesh][mod-mesh]	Supplied 3D geometry or animated meshes	`#E0D0B8`

Benchmark Task identifies the evaluated model family, not its input modality. Both badges indicate evaluation of both families; N/A means unspecified or outside these families.

Task	Evaluated reasoning	Color
![language][task-language]	Reasoning by language models, including multimodal LMs	`#AEC6DF`
![vision][task-vision]	Reasoning by visual generation models	`#C3E6CB`

🧠 Reasoning Engines

🗒️ Grounded Reasoning

Language-led inference, evidence-linked training, distillation, and reliability.

Paper	Resources	Input modalities	Time	Venue
What Do Hallucinations Reveal About Multimodal Reasoning? Diagnosing Visual Grounding Failures via Contrastive Decoding Probes	![resource:code][res-github-code]	![Text][mod-text]	2026-09	`arXiv`
Efficient Reasoning Distillation: Small Video-Language Models via Synthetic CoT and Difficulty-Aware Fine-Tuning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Long-to-Short Video Evidence Reasoning for Grounded Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
BodyCam-VQA: Enhanced Body-Worn Camera Video Captioning via Multimodal Reasoning and Probe Question Generation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Video-MOPD: Multi-Teacher On-Policy Distillation for Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Learning Compositional Spatio-Temporal Video Grounding with Synthetic Curriculum	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Video-OPSD: Exploiting Privileged Visual Evidence for On-Policy Self-Distillation in Video Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Self-Reflective Multi-modal Reasoning for Short-Video Fake News Detection	`N/A`	![Text][mod-text]	2026-08	`arXiv`
Reason in the Words You Speak: Idiolectal Paraphrasing Off-Policy Traces for Reasoning Distillation in VideoLLMs	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Video-FLAIR: Not Whether to Reason, But How	`N/A`	![Text][mod-text]	2026-08	`arXiv`
Finding the Right Evidence: Factor-Guided Coarse-to-Fine Reasoning for Long Videos	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Where to Look Matters: On-Policy Self-Distillation for Long-Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Multi-Agent Self-Improving Reinforcement Learning for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Improving Spatial-Temporal Reasoning in Video-Language Models with Structured Video Prompting	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
COMET: Contrastive Motion-Enhanced Temporal Reasoning for Video Multimodal Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Enhancing Localized Reasoning for Long Video Understanding via Efficient Segment-to-Video Supervision	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Annotations as Rollouts: Efficient and Scalable Reinforcement Learning for Video MLLMs	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Deep Thought Alignment: Trajectory-Level Latent Distillation for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
TennisVAR: A Stroke-Evidence-Grounded Multimodal Large Language Model for Tactical Reasoning in Tennis Videos	![resource:project][res-project]	![Text][mod-text]	2026-08	`arXiv`
AdaThinkV: Adaptive Thinking for Token-Efficient Video Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Adaptive Emotional Video Captioning via Affective Heterogeneous Graph Reasoning and Multi-task Joint Learning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Knowledge-Guided Multimodal Reasoning over Interacting Streams for Video-Level Ambivalence and Hesitancy Recognition	`N/A`	![Text][mod-text]	2026-07	`arXiv`
CADER: Confidence-Aware Dynamic Evidence Reasoning for Long-Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
O-VAD: Industrial Video Anomaly Detection through Object-Centric Tracking and Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
ConsiSpace: Learning Geometric Consistency Matters for Video Spatial Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
GHR-VLM: Making Zero-Shot Transit Video Analytics Realizable with Grounded Hybrid Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Evidence-Backed Video Question Answering	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
TimeThink: Reasoning with Time for Video LLMs	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
EventCoT: Event-centric Video Chain-of-thought for Reasoning Temporal Localization	`N/A`	![Text][mod-text]	2026-07	`arXiv`
STAC: Selective Spatiotemporal Aggregation and Compression for Video Reasoning Segmentation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
EFlow: Learning Evidence Flow for Long-Video Reasoning with Adaptive Reflection	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Linguistic Relative Policy Optimization for Video Anomaly Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Reflect-R1: Evidence-Driven Reflection for Self-Correction in Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
SER: Learning to Ground Video Reasoning with Semantic Evidence Rewards	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
CineCap: Structured Reasoning with Spatio-Temporal Anchors for Cinematographic Video Captioning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
CARE: Competence-Aware Reward Shaping for Adaptive Reasoning Length in Video-MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
APT: Atomic Physical Transitions for Causal Video-Language Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Reasoning as Intersection: Consensus-Frame Alignment for Visual Focus in Video-MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Training LLMs with Reinforcement Learning over Digital Twin Representations for Reasoning-Intensive Surgical VideoQA	`N/A`	![Text][mod-text]	2026-06	`arXiv`
Temporal-Aware Reasoning Optimization for Video Temporal Grounding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Counterfactual Reasoning for Fine-Grained Evidence Disentanglement in VideoQA	`N/A`	![Text][mod-text]	2026-06	`arXiv`
See More, Think Deeper: Query-Expanded Visual Evidence and Answer-Clue Guided Reflection for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
When Video Misreads: Closed-Loop Distillation of Reading Heuristics for Exploratory Manipulation Trace QA	`N/A`	![Text][mod-text]	2026-06	`arXiv`
CACR:Reinforcing Temporal Answer Grounding in Instructional Video via Candidate-Aware Causal Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
VideoSEG-O3: A Multi-turn Reinforcement Learning Framework for Reasoning Video Object Segmentation	![resource:code][res-github-code]	![Text][mod-text]	2026-06	`arXiv`
VLMs are Good Teachers for Video Reasoning via Adaptive Test-Time Optimization	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Question-Aware Evidence Ledgers for Video Relational Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Training-Free Composed Video Retrieval via Visual Representation-Guided Video-LLM Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
TLG: Temporal-Logic Grounding for Video Question Answering via Source-Annotation Reconstruction and Category-Targeted Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video] ![State][mod-state]	2026-06	`arXiv`
Perception First: A Frontier Native-Video Model with Self-Consistency for Implicit Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
R^3: Composed Video Retrieval via Reasoning-Guided Recalling and Re-ranking	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Adaptive Dense Evidence Refinement for Video Relational Reasoning for VRR-QA Challenge	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Disagreement-Based Cross-Model Routing for Implicit Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Reason, Retrieve, Re-rank: A Zero-Shot Reasoning-Aware Framework for Composed Video Retrieval	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Rethinking Video-Language Model from the Language Input Perspective	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Reflective Dialogue between Teacher and Solver Agents for Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Clinically-Grounded Counterfactual Reasoning for Medical Video Diagnosis	`N/A`	![Text][mod-text]	2026-05	`CVPR 2026`
CoReVAD: A Contextual Reasoning Framework for Training-Free Video Anomaly Detection	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Learning Spatiotemporal Sensitivity in Video LLMs via Counterfactual Reinforcement Learning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Foresee-to-Ground: From Predictive Temporal Perception to Evidence-Driven Reasoning for Video Temporal Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
EvoVid: Temporal-Centric Self-Evolution for Video Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
SteerSeg: Attention Steering for Reasoning Video Segmentation	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Video-Zero: Self-Evolution Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Separate First, Fuse Later: Mitigating Cross-Modal Interference in Audio-Visual LLMs Reasoning with Modality-Specific Chain-of-Thought	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-05	`arXiv`
RCoT-Seg: Reinforced Chain-of-Thought for Video Reasoning and Segmentation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
VISD: Enhancing Video Reasoning via Structured Self-Distillation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Think, then Score: Decoupled Reasoning and Scoring for Video Reward Modeling	`N/A`	![Text][mod-text]	2026-05	`arXiv`
From Priors to Perception: Grounding Video-LLMs in Physical Reality	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Beyond Perceptual Shortcuts: Causal-Inspired Debiasing Optimization for Generalizable Video Reasoning in Lightweight MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`CVPR 2026`
Co-Evolving Policy Distillation	`N/A`	![Text][mod-text]	2026-04	`arXiv`
StoryTR: Narrative-Centric Video Temporal Retrieval with Theory of Mind Reasoning	`N/A`	![Text][mod-text]	2026-04	`arXiv`
UpstreamQA: A Modular Framework for Explicit Reasoning on Video Question Answering Tasks	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
ProcessThinker: Enhancing Multi-modal Large Language Models Reasoning via Rollout-based Process Reward	`N/A`	![Text][mod-text]	2026-04	`arXiv`
Seeing Fast and Slow: Learning the Flow of Time in Videos	![resource:project][res-project]	![Text][mod-text]	2026-04	`arXiv`
EasyVideoR1: Easier RL for Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Find, Fix, Reason: Context Repair for Video Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Lost in Adaptation: Layer-Selective Recovery of Temporal Reasoning in Video-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Do Thought Streams Matter? Evaluating Reasoning in Gemini Vision-Language Models for Video Scene Understanding	`N/A`	![Text][mod-text]	2026-04	`arXiv`
OmniJigsaw: Enhancing Omni-Modal Reasoning via Modality-Orchestrated Reordering	![resource:project][res-project]	![Text][mod-text] ![Audio][mod-audio]	2026-04	`arXiv`
Reasoning-Guided Grounding: Elevating Video Anomaly Detection through Multimodal Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
STEER: Structured Event Evidence for Video Reasoning via Multi-Objective Reinforcement Learning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Reinforce to Learn, Elect to Reason: A Dual Paradigm for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`CVPR 2026`
STEAR: Layer-Aware Spatiotemporal Evidence Intervention for Hallucination Mitigation in Video Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
STRIVE: Structured Spatiotemporal Exploration for Reinforcement Learning in Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Reinforcing Consistency in Video MLLMs with Structured Rewards	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
TTA-Vid: Generalized Test-Time Adaptation for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement Learning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Incentivizing Temporal-Awareness in Egocentric Video Understanding Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
VIRST: Video-Instructed Reasoning Assistant for SpatioTemporal Segmentation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
Reinforcing Structured Chain-of-Thought for Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
GridVAD: Open-Set Video Anomaly Detection via Spatial Reasoning over Stratified Frame Grids	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Video-Only ToM: Enhancing Theory of Mind in Multimodal Large Language Models	![resource:project][res-project]	![Text][mod-text]	2026-03	`CVPR 2026`
Learning Trajectory-Aware Multimodal Large Language Models for Video Reasoning Segmentation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
CoVR-R:Reason-Aware Composed Video Retrieval	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Narrative Aligned Long Form Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Learning Transferable Temporal Primitives for Video Reasoning via Synthetic Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
When Thinking Hurts: Mitigating Visual Forgetting in Video Reasoning via Frame Repetition	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Video-CoE: Reinforcing Video Event Prediction via Chain of Events	`N/A`	![Text][mod-text]	2026-03	`CVPR 2026`
VisionCoach: Reinforcing Grounded Video Reasoning via Visual-Perception Prompting	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
SPARROW: Learning Spatial Precision and Temporal Referential Consistency in Pixel-Grounded Video MLLMs	![resource:project][res-project] ![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
Beyond Single-Sample: Reliable Multi-Sample Distillation for Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Are Video Reasoning Models Ready to Go Outside?	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Video-Based Reward Modeling for Computer-Use Agents	`N/A`	![Text][mod-text]	2026-03	`arXiv`
Geometry-Aware Semantic Reasoning for Training Free Video Anomaly Detection	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
SarcasmMiner: A Dual-Track Post-Training Framework for Robust Audio-Visual Sarcasm Reasoning	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-03	`arXiv`
Weakly Supervised Video Anomaly Detection with Anomaly-Connected Components and Intention Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`CVPR 2026`
APPO: Attention-guided Perception Policy Optimization for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`CVPR 2026`
Think with Grounding: Curriculum Reinforced Reasoning with Video Grounding for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
STVG-R1: Incentivizing Instance-Level Reasoning and Grounding in Videos via Reinforcement Learning	`N/A`	![Text][mod-text]	2026-02	`arXiv`
Credit Where It is Due: Cross-Modality Connectivity Drives Precise Reinforcement Learning for MLLM Reasoning	`N/A`	![Text][mod-text]	2026-02	`arXiv`
VideoVeritas: AI-Generated Video Detection via Perception Pretext Reinforcement Learning	![resource:code][res-github-code]	![Text][mod-text]	2026-02	`arXiv`
Process-of-Thought Reasoning for Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
OmniVideo-R1: Reinforcing Audio-visual Reasoning with Query Intention and Modality Attention	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-02	`arXiv`
AVERE: Improving Audiovisual Emotion Reasoning with Preference Optimization	![resource:project][res-project]	![Text][mod-text] ![Audio][mod-audio]	2026-02	`arXiv`
GuardReasoner-Omni: A Reasoning-based Multi-modal Guardrail for Text, Image, Video, and Audio	`N/A`	![Text][mod-text]	2026-02	`arXiv`
Video-OPD: Efficient Post-Training of Multimodal Large Language Models for Temporal Video Grounding via On-Policy Distillation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
RANKVIDEO: Reasoning Reranking for Text-to-Video Retrieval	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
LongVPO: From Anchored Cues to Self-Reasoning for Long-Form Video Preference Optimization	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
SRVAU-R1: Enhancing Video Anomaly Understanding via Reflection-Aware Learning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Structure Over Scale: Learning Visual Reasoning from Pedagogical Video	`N/A`	![Text][mod-text]	2026-01	`arXiv`
Triage: Hierarchical Visual Budgeting for Efficient Video Reasoning in Vision-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Enhancing Video Representations with Spatiotemporal-Semantic Residual to Mitigate Hallucinations in Video Large Multimodal Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning	`N/A`	![Text][mod-text]	2026-01	`arXiv`
Video-KTR: Reinforcing Video Reasoning via Key Token Attribution	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Integrating Fine-Grained Audio-Visual Evidence for Robust Multimodal Emotion Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Audio][mod-audio]	2026-01	`arXiv`
Order from Chaos: Physical World Understanding from Glitchy Gameplay Videos	`N/A`	![Text][mod-text]	2026-01	`arXiv`
Training-Free and Interpretable Hateful Video Detection via Multi-stage Adversarial Reasoning	![resource:code][res-github-code]	![Text][mod-text]	2026-01	`arXiv`
Advancing Adaptive Multi-Stage Video Anomaly Reasoning: A Benchmark Dataset and Method	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
VERHallu: Evaluating and Mitigating Event Relation Hallucination in Video Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
CASHEW: Stabilizing Multimodal Reasoning via Iterative Trajectory Aggregation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Video Evidence to Reasoning Efficient Video Understanding via Explicit Evidence Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-01	`CVPR 2026`
CounterVid: Counterfactual Video Generation for Mitigating Action and Temporal Hallucinations in Video-Language Models	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Analyzing Reasoning Consistency in Large Multimodal Models under Cross-Modal Conflicts	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
PrismVAU: Prompt-Refined Inference System for Multimodal Video Anomaly Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
V-CORE: Temporally Consistent Video Understanding for Video-LLM	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Explicit Abstention Knobs for Predictable Reliability in Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
VideoCuRL: Video Curriculum Reinforcement Learning with Orthogonal Difficulty Decomposition	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`ACL 2026`
Robust Egocentric Referring Video Object Segmentation via Dual-Modal Causal Intervention	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
Factorized Learning for Temporally Grounded Video-Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
SmartSight: Mitigating Hallucination in Video-LLMs Without Compromising Video Understanding via Temporal Attention Collapse	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
Xiaomi MiMo-VL-Miloco Technical Report	![resource:code][res-github-code]	![Text][mod-text]	2025-12	`arXiv`
Rethinking Chain-of-Thought Reasoning for Videos	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
1+1 > 2 : Detector-Empowered Video Large Language Model for Spatio-Temporal Grounding and Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
TempR1: Improving Temporal Understanding of MLLMs via Temporal-Aware Multi-Task Reinforcement Learning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`
OneThinker: All-in-one Reasoning Model for Image and Video	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`
WorldMM: Dynamic Multimodal Memory Agent for Long Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`
Thinking with Drafts: Speculative Temporal Reasoning for Efficient Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-11	`CVPR 2026`
Video-R2: Reinforcing Consistent and Grounded Reasoning in Multimodal Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
Video-CoM: Interactive Video Reasoning via Chain of Manipulations	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
VideoSeg-R1: Reasoning Video Object Segmentation via Reinforcement Learning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
AVATAAR: Agentic Video Answering via Temporal Adaptive Alignment and Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-11	`arXiv`
Agentic Video Intelligence: A Flexible Framework for Advanced Video Exploration and Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
Video Spatial Reasoning with Object-Centric 3D Rollout	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
ViSS-R1: Self-Supervised Reinforcement Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
Video-Thinker: Sparking "Thinking with Videos" via Reinforcement Learning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-10	`arXiv`
Open-o3 Video: Grounded Video Reasoning with Explicit Spatio-Temporal Evidence	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-10	`arXiv`
MOSS-ChatV: Reinforcement Learning with Process Reasoning Reward for Video Temporal Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-09	`arXiv`
VideoChat-R1.5: Visual Test-Time Scaling to Reinforce Multimodal Reasoning by Iterative Perception	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-09	`arXiv`
Strefer: Empowering Video LLMs with Space-Time Referring and Reasoning via Synthetic Instruction Data	![resource:code][res-github-code] weights: Google_Drive	![Text][mod-text] ![Video][mod-video]	2025-09	`arXiv`
Kwai Keye-VL 1.5 Technical Report	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-09	`arXiv`
Video-MTR: Reinforced Multi-Turn Reasoning for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-08	`ICML 2026`
Ovis2.5 Technical Report	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-08	`arXiv`
Reinforcing Video Reasoning Segmentation to Think Before It Segments	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-08	`CVPR 2026`
TAR: Temporal Anchor-Constrained Reasoning for Video Temporal Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-08	`ECCV 2026`
ReasoningTrack: Chain-of-Thought Reasoning for Long-term Vision-Language Tracking	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-08	`arXiv`
Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video Reasoning	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-08	`arXiv`
AVATAR: Reinforcement Learning to See, Hear, and Reason Over Video	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-08	`CVPR 2026`
VideoForest: Person-Anchored Hierarchical Reasoning for Cross-Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-08	`ACM-MM 2025`
ReasonAct: Progressive Training for Fine-Grained Video Reasoning in Small Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-08	`arXiv`
ARC-Hunyuan-Video-7B: Structured Video Comprehension of Real-World Shorts	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-07	`arXiv`
METER: Multi-modal Evidence-based Thinking and Explainable Reasoning -- Algorithm and Benchmark	`N/A`	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-07	`arXiv`
CoTasks: Chain-of-Thought based Video Instruction Tuning Tasks	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-07	`arXiv`
EmbRACE-3K: Embodied Reasoning and Action in Complex Environments	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-07	`arXiv`
ViTCoT: Video-Text Interleaved Chain-of-Thought for Boosting Video Understanding in Large Language Models	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-07	`ACM-MM 2025`
Scaling RL to Long Videos	![resource:code][res-github-code] ![resource:data][res-hf-data] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-07	`NeurIPS 2025`
Video-RTS: Rethinking Reinforcement Learning and Test-Time Scaling for Efficient and Enhanced Video Reasoning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-07	`EMNLP 2025`
Kwai Keye-VL Technical Report	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-07	`arXiv`
Temporal Chain of Thought: Long-Video Understanding by Thinking in Frames	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-07	`arXiv`
DIVE: Deep-search Iterative Video Exploration	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`CVPR 2025`
HumanOmniV2: From Understanding to Omni-Modal Reasoning with Context	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-06	`arXiv`
VLN-R1: Vision-Language Navigation via Reinforcement Fine-Tuning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Ego-R1: Chain-of-Tool-Thought for Ultra-Long Egocentric Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
DAVID-XR1: Detecting AI-Generated Videos with Explainable Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
VideoDeepResearch: Long Video Understanding With Agentic Tool Using	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
VidBridge-R1: Bridging QA and Captioning for RL-based Video Understanding Models with Intermediate Proxy Tasks	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Wait, We Don't Need to "Wait"! Removing Thinking Tokens Improves Reasoning Efficiency	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
DeepVideo-R1: Video Reinforcement Fine-Tuning via Difficulty-aware Regressive GRPO	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`NeurIPS 2025`
VideoChat-A1: Thinking with Long Videos by Chain-of-Shot Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
MiMo-VL Technical Report	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Video-Skill-CoT: Skill-based Chain-of-Thoughts for Domain-Adaptive Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`EMNLP 2025 (Findings)`
EgoVLM: Policy Optimization for Egocentric Video Understanding	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Reinforcement Learning Tuning for VideoLLMs: Reward Design and Data Efficiency	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
VideoCap-R1: Enhancing MLLMs for Video Captioning via Structured Thinking	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
ReAgent-V: A Reward-Driven Multi-Agent Framework for Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-06	`NeurIPS 2025`
ReFoCUS: Reinforcement-guided Frame Optimization for Contextual Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Video-CoT: A Comprehensive Dataset for Spatiotemporal Understanding of Videos Based on Chain-of-Thought	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2025-06	`arXiv`
Chain-of-Frames: Advancing Video Understanding in Multimodal LLMs via Frame-Aware Reasoning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`CVPR 2026`
SiLVR: A Simple Language-based Video Reasoning Framework	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`TMLR 2026`
Reinforcing Video Reasoning with Focused Thinking	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
Fostering Video Reasoning via Next-Event Prediction	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
A2Seek: Towards Reasoning-Centric Benchmark for Aerial Anomaly Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
Omni-R1: Reinforcement Learning for Omnimodal Reasoning via Two-System Collaboration	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-05	`NeurIPS 2025`
Vad-R1: Towards Video Anomaly Reasoning via Perception-to-Cognition Chain-of-Thought	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`NeurIPS 2025`
VerIPO: Cultivating Long Reasoning in Video-LLMs via Verifier-Guided Iterative Policy Optimization	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
Fact-R1: Towards Explainable Video Misinformation Detection with Deep Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-05	`NeurIPS 2025`
Pixel Reasoner: Incentivizing Pixel-Space Reasoning with Curiosity-Driven Reinforcement Learning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`NeurIPS 2025`
UniVG-R1: Reasoning Guided Universal Visual Grounding with Reinforcement Learning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
VideoRFT: Incentivizing Video Reasoning Capability in MLLMs via Reinforced Fine-Tuning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`NeurIPS 2025`
VISTA: Mitigating Semantic Inertia in Video-LLMs via Training-Free Dynamic Chain-of-Thought Routing	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
Seed1.5-VL Technical Report	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
TEMPURA: Temporal Event Masked Prediction and Understanding for Reasoning in Action	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-05	`arXiv`
AVA: Towards Agentic Video Analytics with Vision Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`NSDI 2026`
MR. Video: "MapReduce" is the Principle for Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
Eagle 2.5: Boosting Long-Context Post-Training for Frontier Vision-Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
Multimodal Long Video Modeling Based on Temporal Dynamic Context	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-04	`arXiv`
TinyLLaVA-Video-R1: Towards Smaller LMMs for Video Reasoning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
LVC: A Lightweight Compression Framework for Enhancing VLMs in Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
Spatial-R1: Enhancing MLLMs in Video Spatial Reasoning	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
WikiVideo: Article Generation from Multiple Videos	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
Improved Visual-Spatial Reasoning via R1-Zero-Like Training	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-04	`arXiv`
Aurelia: Test-time Reasoning Distillation in Audio-Visual LLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-03	`ICCV 2025`
Video-R1: Reinforcing Video Reasoning in MLLMs	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-03	`NeurIPS 2025`
VideoMind: A Chain-of-LoRA Agent for Temporal-Grounded Video Reasoning	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-03	`ICLR 2026`
Time-R1: Post-Training Large Vision Language Model for Temporal Video Grounding	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-03	`NeurIPS 2025`
ST-Think: How Multimodal Large Language Models Reason About 4D Worlds from Ego-Centric Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-03	`NeurIPS 2025`
TheoremExplainAgent: Towards Video-based Multimodal Explanations for LLM Theorem Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-02	`ACL 2025 (Oral)`
video-SALMONN-o1: Reasoning-enhanced Audio-visual Large Language Model	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-02	`arXiv`
CoS: Chain-of-Shot Prompting for Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-02	`arXiv`
Temporal Preference Optimization for Long-Form Video Understanding	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-01	`arXiv`
InternLM-XComposer2.5-Reward: A Simple Yet Effective Multi-Modal Reward Model	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-01	`ACL 2025 (Findings)`
MECD+: Unlocking Event-Level Causal Graph Discovery for Video Reasoning	![resource:code][res-github-code] ![resource:weights][res-github-weights]	![Text][mod-text] ![Video][mod-video]	2025-01	`IEEE TPAMI`
Building a Mind Palace: Structuring Environment-Grounded Semantic Graphs for Effective Long Video Analysis with LLMs	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-01	`arXiv`
Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling	![resource:code][res-github-code] ![resource:weights][res-github-weights]	![Text][mod-text] ![Video][mod-video]	2024-12	`arXiv`
STEP: Enhancing Video-LLMs' Compositional Reasoning by Spatio-Temporal Graph-guided Self-Training	`N/A`	![Text][mod-text] ![Video][mod-video]	2024-11	`CVPR 2025`
VideoEspresso: A Large-Scale Chain-of-Thought Dataset for Fine-Grained Video Reasoning via Core Frame Selection	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2024-11	`CVPR 2025`
Adaptive Video Understanding Agent: Enhancing efficiency with dynamic frame sampling and feedback-driven reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2024-10	`NeurIPS 2024 (Workshop)`
VideoINSTA: Zero-shot Long Video Understanding via Informative Spatial-Temporal Reasoning with LLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2024-09	`EMNLP 2024 (Findings)`
MECD: Unlocking Multi-Event Causal Discovery in Video Reasoning	![resource:code][res-github-code] ![resource:weights][res-github-weights]	![Text][mod-text] ![Video][mod-video]	2024-09	`NeurIPS 2024 (Spotlight)`
Video-of-Thought: Step-by-Step Video Reasoning from Perception to Cognition	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2024-05	`ICML 2024`

🎞️ Thinking in Visual Futures

Generated frames and visual updates as intermediate computation, including image-based bridges. Scene time and diffusion steps are distinct.

Paper	Resources	Input modalities	Time	Venue
Do Video Generators Track the World Across Segments? A Benchmark and Method for World-State Reasoning in Video Continuation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Visual prompt engineering for video models	![resource:project][res-project]	![Video][mod-video]	2026-07	`arXiv`
Hierarchical Denoising For Multi-Step Visual Reasoning	![resource:project][res-project]	![Video][mod-video]	2026-07	`arXiv`
OpenCoF: Learning to Reason Through Video Generation	![resource:project][res-project]	![Video][mod-video]	2026-07	`arXiv`
DeltaV: Thinking with Visual State Updates in Unified Large Multimodal Models	![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
Temporal Backtracking Search for Test-time Generative Video Reasoning	`N/A`	![Video][mod-video]	2026-06	`arXiv`
Video Models Can Reason with Verifiable Rewards	`N/A`	![Text][mod-text]	2026-05	`arXiv`
Video Models Reason Early: Exploiting Plan Commitment for Maze Solving	`N/A`	`N/A`	2026-03	`arXiv`
Wan-R1: Verifiable-Reinforcement Learning for Video Reasoning	`N/A`	![Video][mod-video]	2026-03	`arXiv`
Demystifying Video Reasoning	`N/A`	![Video][mod-video]	2026-03	`arXiv`
A Very Big Video Reasoning Suite	![resource:project][res-project]	![Video][mod-video]	2026-02	`arXiv`
Thinking in Frames: How Visual Context and Test-Time Scaling Empower Video Reasoning	`N/A`	![Video][mod-video]	2026-01	`arXiv`
Visual Generation Unlocks Human-Like Reasoning through Multimodal World Models	![resource:project][res-project]	`N/A`	2026-01	`arXiv`
CoF-T2I: Video Models as Pure Visual Reasoners for Text-to-Image Generation	![resource:code][res-github-code]	![Text][mod-text]	2026-01	`arXiv`
VideoCoF: Unified Video Editing with Temporal Reasoner	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Video][mod-video]	2025-12	`CVPR 2026`
McSc: Motion-Corrective Preference Alignment for Video Generation with Self-Critic Hierarchical Reasoning	![resource:code][res-github-code]	`N/A`	2025-11	`arXiv`
In-Video Instructions: Visual Signals as Generative Control	![resource:code][res-github-code]	![Image][mod-image]	2025-11	`arXiv`
Video-as-Answer: Predict and Generate Next Video Event with Joint-GRPO	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text]	2025-11	`arXiv`
Reasoning via Video: The First Evaluation of Video Models’ Reasoning Abilities through Maze-Solving Tasks	![resource:code][res-github-code] ![resource:weights][res-hf-weights] ![resource:data][res-hf-data]	`N/A`	2025-11	`arXiv`
Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm	![resource:code][res-github-code]	`N/A`	2025-11	`arXiv`
Video Models Start to Solve Chess, Maze, Sudoku, Mental Rotation, and Raven’ Matrices	![resource:code][res-github-code]	`N/A`	2025-11	`arXiv`
Are Video Models Ready as Zero-Shot Reasoners? An Empirical Study with the MME-CoF Benchmark	![resource:code][res-github-code] ![resource:data][res-hf-data]	`N/A`	2025-10	`arXiv`
VChain: Chain-of-Visual-Thought for Reasoning in Video Generation	![resource:code][res-github-code]	`N/A`	2025-10	`ACL 2026 (Findings)`

🧩 Stateful Intelligence

Graphs, tracks, spatial maps, and latent reasoning states. Observed memory and imagined state play different roles.

Paper	Resources	Input modalities	Time	Venue
EgoSIS: From Factorized Visual Ego-Transitions to Motion-Canonical Spatial Evidence for UAV Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Interpretable Temporal Video Reasoning with EventGraph and EventField	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Unfold The World: Factorize 4D Properties in Reinforcing Spatial Reasoning	`N/A`	![Text][mod-text]	2026-09	`arXiv`
StateTrace: An Object-Centric Framework for Hidden-State Spatiotemporal Reasoning in Long Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Beyond Visual CoT: Internalized Visual Thinking for Proactive Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
SCOUT: Unlocking Enhanced Spatial Reasoning via Structured Chain-of-Thought and Multi-Objective Process Reward	`N/A`	![Text][mod-text]	2026-08	`arXiv`
R4DSG: Relative 4D Scene Graph Memory for Object-Centric Question Answering in Long Egocentric Video	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Multi-View Relational Distillation for Spatial Reasoning with Vision-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Listen, See and Track: Spatio-Temporal Audio-Visual Sound Event Reasoning for Omni-Modal Language Models	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-08	`arXiv`
Keep It Simple: Multi-Key Episodic Memory Retrieval for Ultra-Long Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
ChronoVision: Temporal Reasoning via Latent State Reconstruction	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
LAVE: Latent Visual Evidence-Enhanced Planning for Video Tool-use Agents	`N/A`	![Video][mod-video]	2026-08	`arXiv`
Perception Before Reasoning: Dynamic Latent Reasoning for Video Understanding and Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Evidence-Grounded Multimodal Knowledge Graph Construction for Multi-Lecture Educational Reasoning	`N/A`	![Text][mod-text]	2026-08	`arXiv`
Beyond Frame Selection: Generative Latent Evidence Aggregation for Long-Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
HyperClaim: Fine-Grained Cross-Modal Hypergraph Reasoning for Video Misinformation Detection	`N/A`	`N/A`	2026-07	`arXiv`
ViSAGE: Constructing Self-Correcting Memories for Long-Form Video Understanding	`N/A`	![Video][mod-video]	2026-07	`arXiv`
ChronoStitch: Training-Free Composition of Visual KV Memories for Long-Horizon Temporal Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
MEMORA: Embodied Action Memory from Egocentric Videos for Reasoning and Planning	![resource:code][res-github-code]	![Video][mod-video]	2026-07	`arXiv`
Whareformer: Learning to Track What is Where in Long Egocentric Videos	![resource:project][res-project]	![Video][mod-video]	2026-07	`arXiv`
Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
SpaceEra++: A Unified Framework Towards 3D Spatial Reasoning in Video	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Latent Visual Cache for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Reasoning with Memory: A Temporal Granularity-Adaptive Framework for Training-Free Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
VisReflect: Latent Visual Reflection for Fine-Grained Perception in Long Visual Context	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
DELTAVID: Enhancing Fine-Grained Spatiotemporal Perception with Cross-Video Differences	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Graph it first! Enabling Reasoning on Long-form Egocentric Videos through Scene Graphs	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
VideoLatent: Video-Language Learning via Latent Self-Forcing	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
ProPlay: Procedural World Models for Self-Evolving LLM Agents	![resource:code][res-github-code]	![Text][mod-text]	2026-06	`arXiv`
From Content to Knowledge: Lightning Fast Long-Video Understanding with Neural Knowledge Representations	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Plan-and-Verify Video Reward Reasoning with Spatio-Temporal Scene Graph Grounding	`N/A`	![Text][mod-text]	2026-06	`arXiv`
Reason, Then Re-reason: Cross-view Revisiting Improves Spatial Reasoning	![resource:project][res-project]	![Text][mod-text]	2026-06	`arXiv`
MemDreamer: Decoupling Perception and Reasoning for Long Video Understanding via Hierarchical Graph Memory and Agentic Retrieval Mechanism	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
MemoryCard: Topic-Aware Multi-Modal Clue Compression for Long-Video Question Answering	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Learning Geometric Representations from Videos for Spatial Intelligent Multimodal Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Imagine Before You Predict: Interleaved Latent Visual Reasoning for Video Event Prediction	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
GOPAgen: Motion-Aware and Efficient Agentic Long-Video Understanding with Structural Memory and Hierarchical Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Reasmory: 3D Reconstruction as Explicit Memory for VLMs Spatial Reasoning	`N/A`	![Text][mod-text]	2026-05	`arXiv`
Q-GeoMem: Question-Guided Geometric Memory for Video Spatial Reasoning	`N/A`	![Text][mod-text]	2026-05	`arXiv`
STORM: Internalized Modeling for Spatial-Temporal Reasoning in Video-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Cambrian-P: Pose-Grounded Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Bernini: Latent Semantic Planning for Video Diffusion	![resource:project][res-project]	![Text][mod-text]	2026-05	`arXiv`
LatentOmni: Rethinking Omni-Modal Understanding via Unified Audio-Visual Latent Reasoning	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-05	`arXiv`
Δynamics: Language-Based Representation for Inferring Rigid-Body Dynamics From Videos	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
LMM-Track4D: Eliciting 4D Dynamic Reasoning in LMMs via Trajectory-Grounded Dialogue	![resource:code][res-github-code]	![Text][mod-text]	2026-05	`arXiv`
PyraVid: Hierarchical Multimodal Memory for Long-Horizon Video Reasoning	`N/A`	![Video][mod-video]	2026-05	`arXiv`
LASAR: Towards Spatio-temporal Reasoning with Latent Cognitive Map	`N/A`	`N/A`	2026-05	`CVPR 2026`
SceneGraphVLM: Dynamic Scene Graph Generation from Video with Vision-Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
SpaceMind++: Toward Allocentric Cognitive Maps for Spatially Grounded Video MLLMs	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Tracking the Truth: Object-Centric Spatio-Temporal Monitoring for Video Large Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Bridging Modalities, Spanning Time: Structured Memory for Ultra-Long Agentic Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Event-Causal RAG: A Retrieval-Augmented Generation Framework for Long Video Reasoning in Complex Scenarios	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
4DThinker: Thinking with 4D Imagery for Dynamic Spatial Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Rethinking Temporal Consistency in Video Object-Centric Learning: From Prediction to Correspondence	`N/A`	`N/A`	2026-05	`arXiv`
Scaling Video Understanding via Compact Latent Multi-Agent Collaboration	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
ONOTE: Hypergraph-Grounded Omnimodal Reasoning for Computational Music Science	`N/A`	`N/A`	2026-04	`arXiv`
Frequency-guided Multi-level Reasoning for Scene Graph Generation in Video	`N/A`	`N/A`	2026-04	`arXiv`
VideoStir: Understanding Long Videos via Spatio-Temporally Structured and Intent-Aware RAG	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-04	`ACL 2026`
Graph-to-Frame RAG: Visual-Space Knowledge Fusion for Training-Free and Auditable Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`CVPR 2026`
A 4D Representation for Training-Free Agentic Reasoning from Monocular Laparoscopic Video	![resource:project][res-project]	![Text][mod-text]	2026-04	`arXiv`
Learning Multi-View Spatial Reasoning from Cross-View Relations	`N/A`	![Text][mod-text]	2026-03	`CVPR 2026`
SpatialStack: Layered Geometry-Language Fusion for 3D VLM Spatial Reasoning	![resource:project][res-project]	![Text][mod-text]	2026-03	`CVPR 2026`
Make Geometry Matter for Spatial Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Unleashing Spatial Reasoning in Multimodal Large Language Models via Textual Representation Guided Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`ACL 2026`
Reconstruction-Guided Slot Curriculum: Addressing Object Over-Fragmentation in Video Object-Centric Learning	`N/A`	`N/A`	2026-03	`CVPR 2026`
Motion-o: Trajectory-Grounded Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Clue Matters: Leveraging Latent Visual Clues to Empower Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Towards Stable Self-Supervised Object Representations in Unconstrained Egocentric Video	`N/A`	![Video][mod-video]	2026-03	`CVPR 2026`
Towards Spatio-Temporal World Scene Graph Generation from Monocular Videos	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
EgoReasoner: Learning Egocentric 4D Reasoning via Task-Adaptive Structured Thinking	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Thinking with Spatial Code for Physical-World Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
FocusGraph: Graph-Structured Frame Selection for Embodied Long Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
ViterbiPlanNet: Injecting Procedural Knowledge via Differentiable Viterbi for Planning in Instructional Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
SCP: Spatial Causal Prediction in Video	![resource:project][res-project]	`N/A`	2026-03	`arXiv`
EgoGraph: Temporal Knowledge Graph for Egocentric Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
TrajTok: Learning Trajectory Tokens enables better Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Keep it SymPL: Symbolic Projective Layout for Allocentric Spatial Reasoning in Vision-Language Models	`N/A`	![Text][mod-text]	2026-02	`CVPR 2026`
GraphThinker: Reinforcing Temporally Grounded Video Reasoning with Event Graph Thinking	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
AD-MIR: Bridging the Gap from Perception to Persuasion in Advertising Video Understanding via Structured Reasoning	![resource:code][res-github-code]	![Video][mod-video]	2026-02	`arXiv`
MosaicThinker: On-Device Visual Spatial Reasoning for Embodied AI via Iterative Construction of Space Representation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Spatial-Conditioned Reasoning in Long-Egocentric Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
SpatialMem: Metric-Aligned Long-Horizon Video Memory for Language Grounding and QA	`N/A`	![Video][mod-video]	2026-01	`arXiv`
MMViR: A Multi-Modal and Multi-Granularity Representation for Long-range Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
AbductiveMLLM: Boosting Visual Abductive Reasoning Within MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Semantic Event Graphs for Long-Form Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Learning to Reason in 4D: Dynamic Spatial Understanding for Vision Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`
4D-RGPT: Toward Region-level 4D Understanding via Perceptual Distillation	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`

💻 Executable Worlds

Executable state, transitions, simulators, and learned rendering. Includes direct video systems and adjacent visual/planning foundations; see the case study.

Paper	Resources	Input modalities	Time	Venue
Recursive Code World Models: Building Complex Worlds through Recursive Scene Programs	`N/A`	![Image][mod-image]	2026-09	`arXiv`
Programmable World Model	![resource:project][res-project] ![resource:code][res-github-code]	![Text][mod-text]	2026-09	`arXiv`
Code as Worlds: Agentic Discovery of Executable World Representations for Physical Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Code World Model: Coding Agent as World Brain	![resource:project][res-project] ![resource:code][res-github-code] ![resource:data][res-hf-data] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Image][mod-image]	2026-08	`arXiv`
Compiling VGDL into Causal Models	`N/A`	![Text][mod-text]	2026-08	`arXiv`
AppDeltaWorld: Transition-Grounded Delta Code World Model for Mobile GUI Agents	`N/A`	`N/A`	2026-08	`arXiv`
PhysMind: From Video to Executable Worlds for Training-Free Physical Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Tycho: Active Abstraction with Programmatic World Models for ARC-AGI-3	![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
VideoCoCo: Code-as-CoT for Physically-Consistent Video Generation via an Agentic Dual-Engine System	`N/A`	![Text][mod-text]	2026-07	`arXiv`
VisualPatchWorld: Code World Models as Latent Structured Representations for Planning	![resource:code][res-github-code]	`N/A`	2026-07	`arXiv`
Agentic Real2Sim: Physics-based World Modeling with Vision-Language Agents	![resource:project][res-project]	![Text][mod-text]	2026-07	`arXiv`
PhysAgent: Reflective Agentic Physics Control for Physically Plausible Video Generation	![resource:project][res-project]	![Text][mod-text]	2026-07	`arXiv`
Do Coding Agents Need Executable World Models, Simplification, and Verification to Solve ARC-AGI-3?	`N/A`	![Text][mod-text]	2026-07	`arXiv`
When a Verified World Model Still Loses: Play-Adequacy vs Prediction-Accuracy in LLM-Synthesized Code World Models	![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
World-Time Compute with Verified Code World Models	![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration for ARC-AGI-3	`N/A`	![Text][mod-text]	2026-07	`arXiv`
World Narrative Model for Highly Controllable Video Generation: A Paradigm Shift from Pixel Sampling to Physical World Orchestration	`N/A`	`N/A`	2026-06	`arXiv`
STRIPS-WM: Learning Grounded Propositional STRIPS-style World Models from Images	`N/A`	![Image][mod-image]	2026-06	`arXiv`
PatchWorld: Gradient-Free Optimization of Executable World Models for Agent Environments	![resource:code][res-github-code]	![Text][mod-text]	2026-05	`arXiv`
Distilling Game Code World Model Generation into Lightweight Large Language Models	`N/A`	![Text][mod-text]	2026-05	`arXiv`
SWoMo: Neuro-Symbolic World Model for Cataract Surgery Simulation	![resource:project][res-project]	`N/A`	2026-05	`arXiv`
Ego2World: Compiling Egocentric Cooking Videos into Executable Worlds for Belief-State Planning	`N/A`	![Video][mod-video]	2026-05	`arXiv`
Executable World Models for ARC-AGI-3 in the Era of Coding Agents	![resource:code][res-github-code]	![Text][mod-text]	2026-05	`arXiv`
Physical Simulator In-the-Loop Video Generation	`N/A`	`N/A`	2026-03	`CVPR 2026`
SimRecon: SimReady Compositional Scene Reconstruction from Real Videos	![resource:project][res-project]	![Video][mod-video]	2026-03	`CVPR 2026`
OOWM: Structuring Embodied Reasoning and Planning via Object-Oriented Programmatic World Modeling	`N/A`	![Text][mod-text]	2026-02	`arXiv`
Neuro-Symbolic Synergy for World Modeling	![resource:code][res-github-code]	![Text][mod-text]	2026-02	`arXiv`
Code2World: A GUI World Model via Renderable Code Generation	![resource:code][res-github-code] ![resource:project][res-project]	![Text][mod-text] ![Action][mod-action]	2026-02	`arXiv`
Generative Visual Code Mobile World Models	`N/A`	![Text][mod-text]	2026-02	`arXiv`
CASSANDRA: Programmatic and Probabilistic Learning and Inference for Stochastic World Modeling	`N/A`	![Text][mod-text]	2026-01	`arXiv`
Agent2World: Learning to Generate Symbolic World Models via Adaptive Multi-Agent Feedback	![resource:project][res-project]	![Text][mod-text]	2025-12	`arXiv`
Generating Code World Models with Large Language Models Guided by Monte Carlo Tree Search	`N/A`	![Text][mod-text]	2024-05	`arXiv`
WorldCoder, a Model-Based LLM Agent: Building World Models by Writing Code and Interacting with the Environment	`N/A`	![Text][mod-text]	2024-02	`arXiv`

🔄 Interaction Loops

🛠️ Active Video Agents

Tool use, evidence search, reflection, and agent coordination. Full-recording retrieval remains an offline access setting.

Paper	Resources	Input modalities	Time	Venue
VideoScout: Learning Agentic Active Exploration with Adaptive Reasoning Pacing for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Online Video Agent Harness for Long Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
One Skill Does Not Fit All: Automatic Discovery and Taxonomy-Guided Routing of Frame-Selection Skills for Long-Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Caption-once, Frames-on-Demand: Visual-Need Routing for Budget-Aware Agentic Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
VLX-VR: An Agentic-Aware Video Reasoning Model	`N/A`	![Video][mod-video]	2026-09	`arXiv`
Thinking with Cameras: Active Visual Reasoning via Dynamic Viewpoint Control for Surveillance Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Visual Search Augmented Chain-of-Thought Reasoning for Attribute Value Extraction from Product Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
From Intent to Evidence: Policy-Steered Multi-Strategy Retrieval for Long-Video Agents	`N/A`	`N/A`	2026-08	`arXiv`
Temporal Tree of Thought: Reasoning-Guided Visual Cue Search for Long-Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
AdaVDR: Adaptive Tool Use and Reflection for Video Deep Research	`N/A`	![Video][mod-video]	2026-08	`arXiv`
VideoHarness-RSI: Recursive Harness Self-Improvement for Long-Video Understanding with Frozen Vision-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Thinking Beyond Videos: Unifying Video Reasoning and Deep Research for Open-World Video Agents	`N/A`	![Video][mod-video]	2026-08	`arXiv`
Coarse Indexing, Fine Evidence: Decoupling Temporal Granularity in Long-Video RAG	`N/A`	![Text][mod-text]	2026-08	`arXiv`
TRACE: Temporal Retrieval with Anchored and Convergent Evidence for Long-Horizon Video Understanding	![resource:project][res-project]	![Video][mod-video]	2026-08	`arXiv`
Routing Before Looking: Query-Adaptive Evidence Acquisition for Long-form Video Understanding	`N/A`	![Video][mod-video]	2026-08	`arXiv`
MedClaw: Heuristic Agent Harness for Long-Horizon Surgical Video Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
REVEAL: A Rubric-Guided Agent for Explicit Evidence Sufficiency Verificationin Long-Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
VTO: Visual Tool Orchestration for Video Anomaly Detection	![resource:code][res-github-code]	![Video][mod-video]	2026-08	`arXiv`
SCOUT: Self-Checking and Recovery-Aware Tool-Thought Agents for Ultra-Long Egocentric Video Reasoning	`N/A`	![Video][mod-video]	2026-08	`arXiv`
SportsGrounder: Proposal-Aided Interleaved Grounding for Dense Sports Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Glance, Scrutinize, and Think: Advancing Video Anomaly Detection from Training-Free to Agentic Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
DAEP: Difficulty-Aware Evidence Planning for Medical Video Corpus Temporal Answer Grounding	`N/A`	`N/A`	2026-08	`arXiv`
Multi-Agent Forensic Reasoning for Generalizable Deepfake Video Detection	![resource:project][res-project]	![Text][mod-text]	2026-08	`arXiv`
MetaVideoAgent: Automated Video-Agent Evolution for Long-Form Video Understanding	![resource:code][res-github-code]	![Video][mod-video]	2026-08	`arXiv`
Video-DeepResearch: Towards the Next-Generation Multimodal Deepresearch Agent	![resource:code][res-github-code]	![Text][mod-text]	2026-08	`arXiv`
When and Where to Look: Adaptive Visual Evidence Scheduling for Efficient Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
AgenticVAU: Multi-Agent Explore-Verify Reasoning for Video Anomaly Understanding	`N/A`	![Video][mod-video]	2026-08	`arXiv`
ReflexTrack: A Feedback-Driven Agent for Training-Free Referring Video Object Segmentation	`N/A`	![Text][mod-text]	2026-07	`arXiv`
MarineEVT: Advancing Event-Centric Marine Video Understanding via Visual Tool Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
OmniReasoner: Thinking with Long Audio-Video via Native Tool Use	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2026-07	`arXiv`
Sparse Evidence Can Suffice: Agentic Evidence Seeking for Multimodal Video Misinformation Detection	`N/A`	`N/A`	2026-07	`arXiv`
Searching Videos as Trees: Self-Correcting Agents for Grounded Long Video QA	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Beyond the Single Camera: Agentic Multi-View Reasoning in Sports Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
TreeSoc: Tree-Structured Dynamic Reasoning and Tool Synergy for Soccer Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
SafeGuard: A Multi-Agent Perception-Reasoning Framework for Social-Risk AI-Generated Video Detection	![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
Incentivizing Vision Language Models to Search for Long Video Question Answering	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
VideoSearcher: Empowering Video Deep Research with Multi-Tool Agentic Reasoning via Reinforcement Learning	![resource:project][res-project] ![resource:code][res-github-code]	![Text][mod-text]	2026-07	`arXiv`
Learning to Watch: Active Video Anomaly Understanding via Interleaved Policy Optimization	`N/A`	![Video][mod-video]	2026-07	`arXiv`
Homer: Understanding Long-form Videos with Hierarchical Memory and Agentic Reasoning	`N/A`	![Text][mod-text]	2026-07	`arXiv`
VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Confidence-Aware Tool Orchestration for Robust Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
video-SALMONN-R³: Learning to ReWatch, ReAsk, and ReAnswer for Efficient Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
VideoAgent: All-in-One Framework for Video Understanding and Editing	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
HPP: Hierarchical Programmatic Probing for Long Video Understanding by Decoupling Perception and Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
TimeProVe: Propose, then Verify for Efficient Long Video Temporal Reasoning in Activities of Daily Living	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence	![resource:project][res-project]	![Text][mod-text]	2026-06	`arXiv`
Native Active Perception as Reasoning for Omni-Modal Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning	`N/A`	`N/A`	2026-06	`arXiv`
Orchestra-o1: Omnimodal Agent Orchestration	`N/A`	![Text][mod-text]	2026-06	`arXiv`
MAVIS: Multi-Agent Video Retrieval via Structured Video Understanding	`N/A`	![Video][mod-video]	2026-06	`ACL 2026 (Findings)`
VTI-CoT: Visual-Textual Interleaved Chain of Thought for Video Reasoning	`N/A`	![Video][mod-video]	2026-06	`arXiv`
AgentCVR: Active Multi-Agent Cross-Video Reasoning via Script-Simulated Reinforcement Learning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Agentic Active Omni-Modal Perception for Multi-Hop Audio-Visual Reasoning	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-05	`arXiv`
DynFrame: Adaptive Reasoning-Driven Multimodal Framework with Dynamic Frame Augmentation for Complex Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Decomposing Queries into Tool Calls for Long-Video Keyframe Retrieval	![resource:code][res-github-code]	![Text][mod-text]	2026-05	`arXiv`
ParaVT: Taming the Tool Prior Paradox for Parallel Tool Use in Agentic Video Reinforcement Learning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Seeing Together: Multi-Robot Cooperative Egocentric Spatial Reasoning with Multimodal Large Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
VideoSeeker: Incentivizing Instance-level Video Understanding via Native Agentic Tool Invocation	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
EvoGround: Self-Evolving Video Agents for Video Temporal Grounding	![resource:project][res-project]	![Video][mod-video]	2026-05	`arXiv`
ReTool-Video: Recursive Tool-Using Video Agents with Meta-Augmented Tool Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
VideoSEAL: Mitigating Evidence Misalignment in Agentic Long Video Understanding by Decoupling Answer Authority	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
ViSRA: A Video-based Spatial Reasoning Agent for Multi-modal Large Language Models	`N/A`	![Text][mod-text]	2026-05	`arXiv`
CollabVR: Collaborative Video Reasoning with Vision-Language and Video Generation Models	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
AffectSeek: Agentic Affective Understanding in Long Videos under Vague User Queries	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
VTAgent: Agentic Keyframe Anchoring for Evidence-Aware Video TextVQA	`N/A`	![Text][mod-text]	2026-05	`arXiv`
Video Active Perception: Effective Inference-Time Long-Form Video Understanding with Vision-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Act2See: Emergent Active Visual Perception for Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`CVPR 2026`
Omni-o3: Deep Nested Omnimodal Deduction for Deliberative Audio-Visual Reasoning	`N/A`	![Audio][mod-audio]	2026-04	`arXiv`
Zoom In, Reason Out: Efficient Far-field Anomaly Detection in Expressway Surveillance Videos via Focused VLM Reasoning Guided by Bayesian Inference	`N/A`	![Text][mod-text]	2026-04	`arXiv`
HiCrew: Hierarchical Reasoning for Long-Form Video Understanding via Question-Aware Multi-Agent Collaboration	`N/A`	![Video][mod-video]	2026-04	`arXiv`
Video-ToC: Video Tree-of-Cue Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
IMPACT-CYCLE: A Contract-Based Multi-Agent System for Claim-Level Supervisory Correction of Long-Video Semantic Memory	![resource:code][res-github-code]	![Video][mod-video]	2026-04	`arXiv`
Chain-of-Glimpse: Search-Guided Progressive Object-Grounded Reasoning for Video Understanding	`N/A`	![Video][mod-video]	2026-04	`arXiv`
Open-Ended Video Game Glitch Detection with Agentic Reasoning and Temporal Grounding	`N/A`	`N/A`	2026-04	`arXiv`
RASR: Retrieval-Augmented Semantic Reasoning for Fake News Video Detection	`N/A`	![Text][mod-text]	2026-04	`arXiv`
SVAgent: Storyline-Guided Long Video Understanding via Cross-Modal Multi-Agent Collaboration	`N/A`	![Video][mod-video]	2026-04	`CVPR 2026`
QVAD: A Question-Centric Agentic Framework for Efficient and Training-Free Video Anomaly Detection	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
Progressive Video Condensation with MLLM Agent for Long-form Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
MAR3: Multi-Agent Recognition, Reasoning, and Reflection for Reference Audio-Visual Segmentation	`N/A`	![Text][mod-text] ![Audio][mod-audio]	2026-03	`arXiv`
VideoTIR: Accurate Understanding for Long Videos with Efficient Tool-Integrated Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
LensWalk: Agentic Video Understanding by Planning How You See in Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
AgentRVOS: Reasoning over Object Tracks for Zero-Shot Referring Video Object Segmentation	![resource:project][res-project]	![Text][mod-text]	2026-03	`arXiv`
EVA: Efficient Reinforcement Learning for End-to-End Video Agent	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
VideoDetective: Clue Hunting via both Extrinsic Query and Intrinsic Relevance for Long Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
VideoSeek: Long-Horizon Video Agent with Tool-Guided Seeking	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
VideoAtlas: Navigating Long-Form Video in Logarithmic Compute	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Interpretable Traffic Responsibility from Dashcam Video via Legal Multi Agent Reasoning	`N/A`	![Text][mod-text]	2026-03	`arXiv`
Symphony: A Cognitively-Inspired Multi-Agent System for Long-Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
A Skill-augmented Agentic Framework and Benchmark for Multi-Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
A Multi-Agent Perception-Action Alliance for Efficient Long Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
Think, Then Verify: A Hypothesis-Verification Multi-Agent Framework for Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`CVPR 2026`
LLandMark: A Multi-Agent Framework for Landmark-Aware Multimodal Interactive Video Retrieval	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Training-Free Spatio-temporal Decoupled Reasoning Video Segmentation with Adaptive Object Memory	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
FactGuard: Agentic Video Misinformation Detection via Reinforcement Learning	`N/A`	![Text][mod-text]	2026-02	`arXiv`
LongVideo-R1: Smart Navigation for Low-cost Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-02	`CVPR 2026`
VIGiA: Instructional Video Guidance via Dialogue Reasoning and Retrieval	`N/A`	![Video][mod-video]	2026-02	`arXiv`
Towards Sparse Video Understanding and Reasoning	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-02	`CVPR 2026`
MedScope: Incentivizing "Think with Videos" for Clinical Reasoning via Coarse-to-Fine Tool Calling	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Agentic Spatio-Temporal Grounding via Collaborative Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
VideoTemp-o3: Harmonizing Temporal Grounding and Video Understanding in Agentic Thinking-with-Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Weaver: End-to-End Agentic System Training for Video Interleaved Reasoning	`N/A`	![Video][mod-video]	2026-02	`arXiv`
VideoBrain: Learning Adaptive Frame Sampling for Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
OmniRAG-Agent: Agentic Omnimodal Reasoning for Low-Resource Long Audio-Video Question Answering	`N/A`	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2026-02	`arXiv`
Refer-Agent: A Collaborative Multi-Agent System with Reasoning and Reflection for Referring Video Object Segmentation	![resource:code][res-github-code]	![Text][mod-text]	2026-02	`CVPR 2026`
Video-o3: Native Interleaved Clue Seeking for Long Video Multi-Hop Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Agentic Very Long Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-01	`ACL 2026`
VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
Hierarchical Long Video Understanding with Audiovisual Entity Cohesion and Agentic Search	`N/A`	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2026-01	`CVPR 2026`
TIR-Flow: Active Video Search and Reasoning with Frozen VLMs	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
FastV-RAG: Towards Fast and Fine-Grained Video QA with Retrieval-Augmented Generation	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`ACL 2026`
Active Perception Agent for Omnimodal Audio-Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-12	`arXiv`
Video-Browser: Towards Agentic Open-web Video Browsing	![resource:code][res-github-code]	![Video][mod-video]	2025-12	`arXiv`
VideoZoomer: Reinforcement-Learned Temporal Focusing for Long Video Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
LongVideoAgent: Multi-Agent Reasoning with Long Videos	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2025-12	`ACL 2026`
Video Detective: Seek Critical Clues Recurrently to Answer Question from Long Videos	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
AdaTooler-V: Adaptive Tool-Use for Images and Videos	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Image][mod-image]	2025-12	`ACL 2026 (Findings)`
LongVT: Incentivizing "Thinking with Long Videos" via Native Tool Calling	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-11	`CVPR 2026`
Video-R4: Reinforcing Text-Rich Video Reasoning with Visual Rumination	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-11	`arXiv`
Orion: A Unified Visual Agent for Multimodal Perception, Advanced Visual Reasoning and Execution	`N/A`	![Text][mod-text]	2025-11	`arXiv`
Video-in-the-Loop: Span-Grounded Long Video QA with Interleaved Reasoning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-10	`arXiv`
FrameMind: Frame-Interleaved Video Reasoning via Reinforcement Learning	`N/A`	![Video][mod-video]	2025-09	`arXiv`
Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video Reasoning	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-08	`arXiv`
ViTCoT: Video-Text Interleaved Chain-of-Thought for Boosting Video Understanding in Large Language Models	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-07	`ACM-MM 2025`

🔁 Unified Understanding & Generation

Architectures combining understanding and generation; iterative reasoning requires separate evidence.

Paper	Resources	Input modalities	Time	Venue
JavisGPT: A Unified Multi-modal LLM for Sounding-Video Comprehension and Generation	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-11	`NeurIPS 2025 (Spotlight)`
VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generation	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text]	2024-09	`ICLR 2025`

📡 Always-On Intelligence

Observed-prefix reasoning, persistent memory, response timing, and proactive interaction.

Paper	Resources	Input modalities	Time	Venue
SVMemAgent: A Streaming Video Memory Agent for Query-Agnostic Online Frame Selection	`N/A`	![Video][mod-video]	2026-09	`arXiv`
Omni-Streaming Thinking	`N/A`	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2026-09	`arXiv`
ReactVAU: A Slow-Fast Decoupled Framework for Streaming Video Anomaly Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Beyond Retrieval: Progressive Latent Memory Evolution for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
CoFiE: Coarse-to-Fine Evidence Selection for Efficient Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
ShallowStream: Index Shallow then Answer Deep for Streaming Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
StreamScout: Learning When to Look Deeper for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Dynamic Hub-and-Spoke Memory for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
StreamEMS: Streaming Video Understanding with Self-Evolving Memory Scheme for Vision-Language Models	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Emotion Understanding in Streaming Video with Trajectory-Aware Reliability	`N/A`	![Video][mod-video]	2026-08	`arXiv`
Strictly Causal Streaming Video Anomaly Detection with a Theoretically-Grounded State-Space Core	`N/A`	![Video][mod-video]	2026-08	`arXiv`
StreamOPD: A Post-Training Recipe with Spatio-Temporal Cue Gating for Streaming Video Understanding	![resource:project][res-project]	![Video][mod-video]	2026-08	`arXiv`
StreamFlow: Dynamic Memory Flows for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
GROVE: Growing and Reasoning over Temporally Stratified Memory from Streaming Video Experience	![resource:code][res-github-code]	![Video][mod-video]	2026-08	`arXiv`
Think in Sets for Streaming Video Token Compression	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
ObjectStream: Latent Objects as Memory Anchors for Streaming Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Memory-Augmented Multimodal Large Language Models for Small Object Understanding in Streaming Aerial Videos	`N/A`	![Text][mod-text]	2026-07	`arXiv`
FOLIO: Focused Semantic Memory for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
Vinci2: Providing Proactive Assistance in Continuous Egocentric Videos	![resource:project][res-project]	![Video][mod-video]	2026-07	`arXiv`
Light-Omni: Reflex over Reasoning in Agentic Video Understanding with Long-Term Memory	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
ReflectWorld-MM: An Entity-Oriented Multimodal Memory System for Open-Ended Video Streams	`N/A`	![Video][mod-video]	2026-07	`arXiv`
GuideMe: Multi-Domain Task Guidance and Intervention in Streaming Video	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-07	`arXiv`
ProtoKV: Streaming Video Understanding under Delayed Query with Summary-State Memory	`N/A`	![Video][mod-video]	2026-06	`arXiv`
Towards a Dynamic and Fixed-budget Memory Bank for Efficient Streaming Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
ESTANet: Efficient Online Error Detection in Procedural Videos via Prediction Inconsistency	`N/A`	`N/A`	2026-06	`arXiv`
CoVStream: Edge-Cloud Collaboration for Understanding of Long Video Streams	`N/A`	![Video][mod-video]	2026-06	`arXiv`
LiveStarPro: Proactive Streaming Video Understanding with Hierarchical Memory for Long-Horizon Streams	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
What Should a Streaming Video Model Remember?	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Keep It in Mind: User Centric Continual Spatial Intelligence Reasoning in Egocentric Video Streams	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
AdaSR: Adaptive Streaming Reasoning with Hierarchical Relative Policy Optimization	![resource:code][res-github-code]	![Text][mod-text] ![Audio][mod-audio]	2026-06	`arXiv`
DuplexOmni: Real-Time Listening, Seeing, Thinking, and Speaking for Full-Duplex Interaction	`N/A`	![Video][mod-video] ![Audio][mod-audio]	2026-06	`arXiv`
Harnessing Streaming Video in the Wild	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
Don't Pause: Streaming Video-Language Synchrony for Online Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
MOSS-Video-Preview: Toward Real-Time Video Understanding via Cross-Attention	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-06	`arXiv`
FlowNar: Scalable Streaming Narration for Long-Form Videos	![resource:code][res-github-code]	![Text][mod-text]	2026-05	`arXiv`
OmniMem: Perturbation-aware Memory Compression for Streaming Audio-Visual LLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2026-05	`arXiv`
StreamOV: Streaming Omni-Video Understanding via Evidence-Guided Memory and Response Triggering	`N/A`	![Video][mod-video] ![Audio][mod-audio]	2026-05	`arXiv`
MuKV: Multi-Grained KV Cache Compression for Long Streaming Video Question-Answering	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`CVPR 2026`
An Efficient Streaming Video Understanding Framework with Agentic Control	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
CoRDS: Coreset-based Representative and Diverse Selection for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
EvoStreaming: Your Offline Video Model Is a Natively Streaming Assistant	`N/A`	![Text][mod-text]	2026-05	`arXiv`
StreamPro: From Reactive Perception to Proactive Decision-Making in Streaming Video	`N/A`	![Video][mod-video]	2026-05	`arXiv`
Semantic-Aware Adaptive Visual Memory for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Response-G1: Explicit Scene Graph Modeling for Proactive Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`ACL 2026`
Decouple and Cache: KV Cache Construction for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-05	`arXiv`
Progressive Online Video Understanding with Evidence-Aligned Timing and Transparent Decisions	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
OASIS: On-Demand Hierarchical Event Memory for Streaming Video Reasoning	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-04	`CVPR 2026`
Online Reasoning Video Object Segmentation	`N/A`	`N/A`	2026-04	`arXiv`
StreamMeCo: Long-Term Agent Memory Compression for Efficient Streaming Video Understanding	![resource:code][res-github-code]	![Video][mod-video]	2026-04	`ACL 2026 (Findings)`
ESOM: Efficiently Understanding Streaming Video Anomalies with Open-world Dynamic Definitions	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference	`N/A`	![Text][mod-text]	2026-04	`arXiv`
AURA: Always-On Understanding and Real-Time Assistance via Video Streams	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
A Simple Baseline for Streaming Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-04	`arXiv`
STRIDE: When to Speak Meets Sequence Denoising for Streaming Video Understanding	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
GazeQwen: Lightweight Gaze-Conditioned LLM Modulation for Streaming Video Understanding	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
StreamingClaw Technical Report	`N/A`	![Video][mod-video]	2026-03	`arXiv`
PEARL: Personalized Streaming Video Understanding Model	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
CurveStream: Boosting Streaming Video Understanding in MLLMs via Curvature-Aware Hierarchical Visual Memory Management	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Em-Garde: A Propose-Match Framework for Proactive Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Thinking in Streaming Video	![resource:code][res-github-code]	![Video][mod-video]	2026-03	`arXiv`
Video Streaming Thinking: VideoLLMs Can Watch and Think Simultaneously	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`ECCV 2026`
WAT: Online Video Understanding Needs Watching Before Thinking	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
Think While Watching: Online Streaming Segment-Level Memory for Multi-Turn Video Reasoning in Multimodal Large Language Models	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`arXiv`
StreamReady: Learning What to Answer and When in Long Streaming Videos	`N/A`	![Video][mod-video]	2026-03	`CVPR 2026`
Proact-VL: A Proactive VideoLLM for Real-Time AI Companions	`N/A`	![Text][mod-text]	2026-03	`arXiv`
Think-as-You-See: Streaming Chain-of-Thought Reasoning for Large Vision-Language Models	![resource:code][res-github-code]	![Text][mod-text]	2026-03	`CVPR 2026`
FluxMem: Adaptive Hierarchical Memory for Streaming Video Understanding	`N/A`	![Video][mod-video]	2026-03	`CVPR 2026`
From Verbatim to Gist: Distilling Pyramidal Multimodal Memory via Semantic Information Bottleneck for Long-Horizon Video Agents	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-03	`ACL 2026`
RT-NeuS: Towards Real-Time Neuro-Symbolic Video Understanding via Adaptive Temporal Verification	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
WeaveTime: Stream from Earlier Frames into Emergent Memory in VideoLLMs	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-02	`CVPR 2026`
Going Down Memory Lane: Scaling Tokens for Video Stream Understanding with Dynamic KV-Cache Memory	![resource:project][res-project]	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
EventMemAgent: Hierarchical Event-Centric Memory for Online Video Understanding with Adaptive Tool Use	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Vista: Scene-Aware Optimization for Streaming Video Question Answering under Post-Hoc Queries	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
FreshMem: Brain-Inspired Frequency-Space Hybrid Memory for Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-02	`arXiv`
Event-VStream: Event-Driven Real-Time Understanding for Long Video Streams	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`arXiv`
HERMES: KV Cache as Hierarchical Memory for Efficient Streaming Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-01	`ACL 2026`
ROMA: Real-time Omni-Multimodal Assistant with Interactive Streaming Understanding	![resource:project][res-project]	![Text][mod-text]	2026-01	`ACL 2026 (Findings)`
Streaming Video Instruction Tuning	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-12	`CVPR 2026`
VideoScaffold: Elastic-Scale Visual Hierarchies for Streaming Video Understanding in MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-12	`arXiv`
LiveStar: Live Streaming Assistant for Real-World Online Video Understanding	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-11	`NeurIPS 2025`
StreamingCoT: A Dataset for Temporal Dynamics and Multimodal Chain-of-Thought Reasoning in Streaming VideoQA	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-10	`ACM-MM 2025`
StreamingVLM: Real-Time Understanding for Infinite Video Streams	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-10	`ICLR 2026`
StreamForest: Efficient Online Video Understanding with Persistent Event Memory	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-09	`NeurIPS 2025 (Spotlight)`
StreamAgent: Towards Anticipatory Agents for Streaming Video Understanding	`N/A`	![Video][mod-video]	2025-08	`arXiv`
StreamVLN: Streaming Vision-and-Language Navigation via SlowFast Context Modeling	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text]	2025-07	`ICRA 2026`
Flash-VStream: Efficient Real-Time Understanding for Long Video Streams	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-06	`ICCV 2025`
LiveVLM: Efficient Online Video Understanding via Streaming-Oriented KV Cache and Retrieval	`N/A`	![Text][mod-text] ![Video][mod-video]	2025-05	`DAC 2026`
StreamBridge: Turning Your Offline Video Large Language Model into a Proactive Streaming Assistant	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-05	`NeurIPS 2025`
TimeChat-Online: 80% Visual Tokens are Naturally Redundant in Streaming Videos	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-04	`ACM-MM 2025`
LiveCC: Learning Video LLM with Streaming Speech Transcription at Scale	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video] ![Audio][mod-audio]	2025-04	`CVPR 2025`
ViSpeak: Visual Instruction Feedback in Streaming Videos	![resource:code][res-github-code] ![resource:weights][res-github-weights]	![Text][mod-text] ![Video][mod-video]	2025-03	`ICCV 2025`
StreamMind: Unlocking Full Frame Rate Streaming Video Dialogue through Event-Gated Cognition	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-03	`ICCV 2025`
Streaming Video Question-Answering with In-context Video KV-Cache Retrieval	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2025-03	`ICLR 2025`
SVBench: A Benchmark with Temporal Multi-Turn Dialogues for Streaming Video Understanding	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-02	`ICLR 2025 (Spotlight)`
Streaming Video Understanding and Multi-round Interaction with Memory-enhanced Knowledge	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2025-01	`ICLR 2025`
Dispider: Enabling Video LLMs with Active Real-Time Interaction via Disentangled Perception, Decision, and Reaction	![resource:code][res-github-code] ![resource:weights][res-hf-weights]	![Text][mod-text] ![Video][mod-video]	2025-01	`CVPR 2025`
Online Video Understanding: OVBench and VideoChat-Online	![resource:code][res-github-code] ![resource:data][res-hf-data]	![Text][mod-text] ![Video][mod-video]	2024-12	`CVPR 2025`
StreamChat: Chatting with Streaming Video	`N/A`	![Text][mod-text] ![Video][mod-video]	2024-12	`arXiv`

🔎 Evidence & Action

🔎 Evidence at Scale

Frame selection, token compression, temporal grounding, and audio-visual representations supporting reasoning.

Paper	Resources	Input modalities	Time	Venue
VideoMM: Adaptive Macro-Micro Inference for Efficient Video MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
MarKey: Marginal Utility Guided Greedy Keyframe Selection for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
IAE-VTG: Interaction-Aligned Action-Entity Video Temporal Grounding	`N/A`	![Video][mod-video]	2026-09	`arXiv`
DSE-VTG: Dual-Side Enhancement for Training-Free Video Temporal Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
Select, Compress, Reinvest: A Controlled Study of Visual-Token Allocation in Long-Video MLLMs	![resource:code][res-github-code]	![Text][mod-text] ![Video][mod-video]	2026-09	`arXiv`
RIDGE: Region-Informed Derivative-Guided Evidence Selection for Long Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Locate Anything in Videos: Rethinking Efficient Generative Spatio-Temporal Video Grounding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
Training-Free Temporal Abstraction for General Video Understanding	`N/A`	![Text][mod-text] ![Video][mod-video]	2026-08	`arXiv`
[LongVU-TTT: Causal Test-Time Training for Visual Resampling in Long Video Understanding](h

Truncated — view the full README on GitHub.

artificial-intelligence

awesome

benchmarks

chain-of-frames

chain-of-thought

large-language-models

multimodal-large-language-models

omni

paper-list

real-time-video-understanding

streaming-video-reasoning

streaming-video-understanding

think-with-video

video-generation

video-reasoning

world-model

Similar repos

yunlong10/Awesome-Video-LMM-Post-Training

🔥🔥🔥 Latest Papers, Codes and Datasets on Video-LMM Post-Training

303

yeliudev/VideoMind

🧠 VideoMind: A Chain-of-LoRA Agent for Temporal-Grounded Video Reasoning (ICLR 2026)

357

The-Martyr/Awesome-Multimodal-Reasoning

Latest Advances on (RL based) Multimodal Reasoning and Generation in Multimodal LLMs

102

HITsz-TMG/Awesome-Large-Multimodal-Reasoning-Models

The development and future prospects of large multimodal reasoning models.

616

lwpyh/Awesome-MLLM-Reasoning-Collection

A collection of multimodal reasoning papers, codes, datasets, benchmarks and resources.

fnlp/VideoThinkBench

Dataset

[CVPR 2026] Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm

yaotingwangofficial/Awesome-MCoT

Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey

1,026

AIWorldLab/Awesome-Vision-World-Model

From Seeing to Knowing the World: A Survey of Vision World Models

193

Contributors

119 commits

18 commits

3 commits

2 commits

Languages

Python

95.8%

JavaScript

3.6%