Awesome Robot Diffusion

A curated list of recent robot learning papers incorporating diffusion models for manipulation, navigation, planning etc. The paper list is structured so that each paper falls into only one place. While some methods could fit into multiple places, we place each one in the most relevant class.

Please refer to our survey paper below for detailed review, which has been accepted to International Journal of Computer Vision (IJCV).

If our work helps you in your research, please kindly cite our paper 😀.

@article{liu2026diffusion,
  title={Diffusion models in robotics: A survey},
  author={Liu, Xiaokang and Ma, Kevin Yuchen and Gao, Chen and Shou, Mike Zheng},
  journal={International Journal of Computer Vision},
  volume={134},
  number={6},
  pages={299},
  year={2026},
  publisher={Springer}
}

---

Table of Contents

• Benchmarks
• Diffusion as Policy
• Diffusion as Synthesizer
• Task Objectives and Applications
• Robot Learning Utilizing Diffusion Model Properties
• Citation

Benchmarks

• Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations (RSS 2018)

• Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning (CoRL 2020)

• Bridge data: Boosting generalization of robotic skills with cross-domain datasets (RSS 2022)

• Towards Human-Level Bimanual Dexterous Manipulation with Reinforcement Learning (NeurIPS 2022 Datasets and Benchmarks Track)

• Dexart: Benchmarking generalizable dexterous manipulation with articulated objects (CVPR 2023)

• BridgeData V2: A Dataset for Robot Learning at Scale (CoRL 2023)

• CALVIN: A Benchmark for Language-conditioned Policy Learning for Long-horizon Robot Manipulation Tasks (RAL 2022)

• RLBench: The Robot Learning Benchmark & Learning Environment (RAL 2020)

• LIBERO: Benchmarking Knowledge Transfer in Lifelong Robot Learning (NeurIPS 2023 Dataset and Benchmark Track)

Diffusion as Policy

• Diffusion policy: Visuomotor policy learning via action diffusion (RSS 2023)

Algorithmic Level (Diffusion Variations)

SGM

• Goal-conditioned imitation learning using score-based diffusion policies (RSS 2023)

• Movement Primitive Diffusion: Learning Gentle Robotic Manipulation of Deformable Objects

• Precise Pick-and-Place using Score-Based Diffusion Networks

Consistency Model

• Consistency policy: Accelerated visuomotor policies via consistency distillation (RSS 2024)

• ManiCM: Real-time 3D Diffusion Policy via Consistency Model for Robotic Manipulation

Flow Matching

• Vision-Language-Affordance-based Robot Manipulation with Flow Matching (Sep 2024)

• Flow Matching Imitation Learning for Multi-Support Manipulation

• FlowPolicy: Enabling Fast and Robust 3D Flow-based Policy via Consistency Flow Matching for Robot Manipulation

• ACTIONFLOW: Equivariant, Accurate, and Efficient Policies with Spatially Symmetric Flow Matching

• Learning Robotic Manipulation Policies from Point Clouds with Conditional Flow Matching

• Fast and Robust Visuomotor Riemannian Flow Matching Policy

• Safe Flow Matching: Robot Motion Planning with Control Barrier Functions

Architecture Level

DiT

• RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation (Dec 2024)

• Diffusion Transformer Policy

• Scaling Diffusion Policy in Transformer to 1 Billion Parameters for Robotic Manipulation

• Prediction with Action: Visual Policy Learning via Joint Denoising Process

• The Ingredients for Robotic Diffusion Transformers

Equivariant Architecture

• EquiBot: SIM (3)-Equivariant Diffusion Policy for Generalizable and Data Efficient Learning (CoRL 2024)

• Equivariant Diffusion Policy (CoRL 2024)

• Diffusion-EDFs: Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulation

• ET-SEED: Efficient Trajectory-Level SE(3) Equivariant Diffusion Policy

• Equivariant Visual-Tactile Diffusion Policy for Contact-Rich Manipulation (CoRL 2026)

Mamba Models

• MaIL: Improving Imitation Learning with Mamba

• Mamba Policy: Towards Efficient 3D Diffusion Policy with Hybrid Selective State models

Hierarchical Design

• SkillDiffuser: Interpretable Hierarchical Planning via Skill Abstractions in Diffusion-Based Task Execution (CVPR 2024)

• Hierarchical Diffusion Policy for Kinematics-Aware Multi-Task Robotic Manipulation (CVPR 2024)

• ChainedDiffuser: Unifying Trajectory Diffusion and Keypose Prediction for Robotic Manipulation (CoRL 2023)

• Hierarchical Diffusion Policy: manipulation trajectory generation via contact guidance

• XSkill: Cross Embodiment Skill Discovery (CoRL 2023)

• Diffskill: Improving Reinforcement Learning Through Diffusion-Based Skill Denoiser for Robotic Manipulation (Knowledge-Based Systems 2024)

• Scaling Up and Distilling Down: Language-Guided Robot Skill Acquisition (CoRL 2023)

• Generative Skill Chaining: Long-Horizon Skill Planning with Diffusion Models (CoRL 2023)

Module Level

Perception: 3D

• 3D Diffuser Actor: Policy Diffusion with 3D Scene Representations (CoRL 2024)

• 3D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations (RSS 2024)

• RISE: 3D Perception Makes Real-World Robot Imitation Simple and Effective (IROS 2024)

• GenDP: 3D Semantic Fields for Category-Level Generalizable Diffusion Policy (CoRL 2024)

• DTG : Diffusion-based Trajectory Generation for Mapless Global Navigation

Perception: Tactile/Force

• 3D-ViTac: Learning Fine-Grained Manipulation with Visuo-Tactile Sensing

• VITaL Pretraining: Visuo-Tactile Pretraining for Tactile and Non-Tactile Manipulation Policies

• Learning Visuotactile Skills with Two Multifingered Hands

• Adaptive Compliance Policy: Learning Approximate Compliance for Diffusion Guided Control

• ForceMimic: Force-Centric Imitation Learning with Force-Motion Capture System for Contact-Rich Manipulation

• Learning Diffusion Policies from Demonstrations For Compliant Contact-rich Manipulation

• TacDiffusion: Force-domain Diffusion Policy for Precise Tactile Manipulation

• Admittance Visuomotor Policy Learning for General-Purpose Contact-Rich Manipulations

• FoAR: Force-Aware Reactive Policy for Contact-Rich Robotic Manipulation

• Canonical Representation and Force-Based Pretraining of 3D Tactile for Dexterous Visuo-Tactile Policy Learning

• Reactive Diffusion Policy: Slow-Fast Visual-Tactile Policy Learning for Contact-Rich Manipulation

Image-goal Conditioning

• Subgoal Diffuser: Coarse-to-fine Subgoal Generation to Guide Model Predictive Control for Robot Manipulation

• Goal-Reaching Policy Learning from Non-Expert Observations via Effective Subgoal Guidance

• GHIL-Glue: Hierarchical Control with Filtered Subgoal Images

• Look Before You Leap: Unveiling the Power of GPT-4V in Robotic Vision-Language Planning

• Track2Act: Predicting Point Tracks from Internet Videos enables Generalizable Robot Manipulation (ECCV 2024)

Language-goal Conditioning

• π0: A Vision-Language-Action Flow Model for General Robot Control

• Multimodal Diffusion Transformer: Learning Versatile Behavior from Multimodal Goals

• PlayFusion: Skill Acquisition via Diffusion from Language-Annotated Play (CoRL 2023)

• Language Control Diffusion: Efficiently Scaling through Space, Time, and Tasks

• DISCO: Language-Guided Manipulation with Diffusion Policies and Constrained Inpainting

• Diffusion-VLA: Scaling Robot Foundation Models via Unified Diffusion and Autoregression (Dec 2024)

• Vision-Language-Action Model and Diffusion Policy Switching Enables Dexterous Control of an Anthropomorphic Hand

• TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation

• VLA-Cache: Towards Efficient Vision-Language-Action Model via Adaptive Token Caching in Robotic Manipulation

Object-centric

• Language-Guided Object-Centric Diffusion Policy for Collision-Aware Robotic Manipulation

• Se(3)-diffusionfields: Learning cost functions for joint grasp and motion optimization through diffusion (ICRA 2023)

• Affordance-Centric Policy Learning: Sample Efficient and Generalisable Robot Policy Learning using Affordance-Centric Task Frames

• Motion Before Action: Diffusing Object Motion as Manipulation Condition

• SPOT: SE(3) Pose Trajectory Diffusion for Object-Centric Manipulation

• G3Flow: Generative 3D Semantic Flow for Pose-aware and Generalizable Object Manipulation

• Composable Part-Based Manipulation (CoRL 2023)

• DAP: Diffusion-based Affordance Prediction for Multi-modality Storage

• StructDiffusion: Language-Guided Creation of Physically-Valid Structures using Unseen Objects

• Shelving, stacking, hanging: Relational pose diffusion for multi-modal rearrangement (CoRL 2023)

• DiffusionNOCS: Managing Symmetry and Uncertainty in Sim2Real Multi-Modal Category-level Pose Estimation

• Reorientdiff: Diffusion model based reorientation for object manipulation (ICRA 2024)

Policy Generalizablity

• CAGE: Causal Attention Enables Data-Efficient Generalizable Robotic Manipulation (single task generalization)

• Planning-Guided Diffusion Policy Learning for Generalizable Contact-Rich Bimanual Manipulation (Sep 2024)

• Learning Generalizable 3D Manipulation With 10 Demonstrations

• JUICER: Data-Efficient Imitation Learning for Robotic Assembly

• Diffusion-PbD: Generalizable Robot Programming by Demonstration with Diffusion Features

• Imitation Learning with Limited Actions via Diffusion Planners and Deep Koopman Controllers

Diffusion as Synthesizer

Planning

• SAGE Improving Diffusion Planners by Self-Supervised Action Gating with Energies (ICML 2026)

• DriveVA: DriveVA: Video Action Models are Zero-Shot Drivers (ECCV 2026)

• DALL-E-Bot: DALL-E-Bot: Introducing Web-Scale Diffusion Models to Robotics (RA-Letters 2023)

• UniPi: Learning Universal Policies via Text-Guided Video Generation (NeurIPS 2023)

• AVDC: Learning to Act from Actionless Videos through Dense Correspondences (ICLR 2024)

• UniSim: UniSim: Learning Interactive Real-World Simulators (ICLR 2024)

• HiP: Compositional Foundation Models for Hierarchical Planning (NeurIPS 2023)

• VLP: Video language planning (ICLR 2024)

• Dreamitate: Dreamitate: Real-World Visuomotor Policy Learning via Video Generation (CoRL 2024)

• ARDuP: ARDuP: Active Region Video Diffusion for Universal Policies (Jun 2024)

• This&That: This&That: Language-Gesture Controlled Video Generation for Robot Planning (July 2024)

• RoboDreamer: RoboDreamer: Learning Compositional World Models for Robot Imagination (ICML 2024)

• CLOVER: Closed-Loop Visuomotor Control with Generative Expectation for Robotic Manipulation (NeurIPS 2024)

• SOAR: Autonomous Improvement of Instruction Following Skills via Foundation Models (CoRL 2024) (augmentation)

• 3D-VLA: A 3D Vision-Language-Action Generative World Model

• IRASim: Learning Interactive Real-Robot Action Simulators

• ViNT: A foundation model for visual navigation

• GR-MG: Leveraging Partially-Annotated Data via Multi-Modal Goal-Conditioned Policy

• IMAGINATION POLICY: Using Generative Point Cloud Models for Learning Manipulation Policies

• Imagine2Servo: Intelligent Visual Servoing with Diffusion-Driven Goal Generation for Robotic Tasks

Data Collection

• DMD: Diffusion Meets DAgger: Supercharging Eye-in-hand Imitation Learning (Feb 2024)

• Cacti: Cacti: A framework for scalable multi-task multi-scene visual imitation learning (CoRL 2022 Workshop PRL)

• GenAug: GenAug: Retargeting behaviors to unseen situations via Generative Augmentation

• Scaling Robot Learning with Semantically Imagined Experience

• Embodiment-agnostic Action Planning via Object-Part Scene Flow

Task Objectives and Applications

Constrained Motion Planning

• EDMP: Ensemble-of-costs-guided Diffusion for Motion Planning (ICRA 2024)

• Spatial-Temporal Graph Diffusion Policy with Kinematic Modeling for Bimanual Robotic Manipulation (CVPR 2025)

• Motion Planning Diffusion: Learning and Adapting Robot Motion Planning with Diffusion Models

• SafeDiffuser: Safe Planning with Diffusion Probabilistic Models

• Sampling Constrained Trajectories Using Composable Diffusion Models

• Constrained Stein Variational Trajectory optimization

• DiMSam: Diffusion Models as Samplers for Task and Motion Planning under Partial Observability

• DiffusionSeeder: Seeding Motion Optimization with Diffusion for Rapid Motion Planning

• LTLDoG: Satisfying Temporally-Extended Symbolic Constraints for Safe Diffusion-based Planning

• Potential Based Diffusion Motion Planning

• RAIL: Reachability-Aided Imitation Learning for Safe Policy Execution

• Learning Diverse Robot Striking Motions with Diffusion Models and Kinematically Constrained Gradient Guidance

Dexterous Manipulation

• ViViDex: Learning Vision-based Dexterous Manipulation from Human Videos

• Vision-Language-Action Model and Diffusion Policy Switching Enables Dexterous Control of an Anthropomorphic Hand

• Learning score-based grasping primitive for human-assisting dexterous grasping (NeurIPS 2023)

• DexDiffuser: Generating Dexterous Grasps with Diffusion Models (Feb 2024)

• Dexterous Functional Pre-Grasp Manipulation with Diffusion Policy (Mar 2024)

• DexGrasp-Diffusion: Diffusion-based Unified Functional Grasp Synthesis Method for Multi-Dexterous Robotic Hands

• Grasp Diffusion Network: Learning Grasp Generators from Partial Point Clouds with Diffusion Models in SO(3) × R3

• PianoMime: Learning a Generalist, Dexterous Piano Player from Internet Demonstrations

• Learning Playing Piano with Bionic-Constrained Diffusion Policy for Anthropomorphic Hand

Human-Robot Interaction

• Diffusion Co-Policy for Synergistic Human-Robot Collaborative Tasks

• Maximizing Alignment with Minimal Feedback: Efficiently Learning Rewards for Visuomotor Robot Policy Alignment

• Task-based dialogue policy learning based on diffusion models

• Autonomous Human-Robot Interaction via Operator Imitation

Mobile Manipulation

• UMI on Legs: Making Manipulation Policies Mobile with Manipulation-Centric Whole-body Controllers (CoRL 2024)

• Pushing the Limits of Cross-Embodiment Learning for Manipulation and Navigation

• M2Diffuser: Diffusion-based Trajectory Optimization for Mobile Manipulation in 3D Scenes

• Robust Imitation Learning for Mobile Manipulator Focusing on Task-Related Viewpoints and Regions

Deformable Objects

• Movement Primitive Diffusion: Learning Gentle Robotic Manipulation of Deformable Objects

• SculptDiff: Learning Robotic Clay Sculpting from Humans with Goal Conditioned Diffusion Policy

• RoPotter: Toward Robotic Pottery and Deformable Object Manipulation with Structural Priors

• Garment Diffusion Models for Robot-Assisted Dressing

Navigation

• NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration

• Diffusion-based 2D Path Planner applied on Legged Robots

• LDP: A Local Diffusion Planner for Efficient Robot Navigation and Collision Avoidance

• DARE: Diffusion Policy for Autonomous Robot Exploration

• FloNa: Floor Plan Guided Embodied Visual Navigation

• Learning Wheelchair Tennis Navigation from Broadcast Videos with Domain Knowledge Transfer and Diffusion Motion Planning

• Learning Efficient Navigation Policies via Conditional Flow Matching

[unclassified]

Generative Image as Action Models

Large-Scale Actionless Video Pre-Training via Discrete Diffusion for Efficient Policy Learning

Diffusion model is an effective planner and data synthesizer for multi-task reinforcement learning

Learning Visual Parkour from Generated Images

VidMan: Exploiting Implicit Dynamics from Video Diffusion Model for Effective Robot Manipulation (inverse dynamics, distillation from video to action generation)

• Imitating Human Behaviour with Diffusion Models (ICLR 2023)

• Memory-Consistent Neural Networks for Imitation Learning (ICLR 2024)

• Differentiable Robot Rendering (CoRL 2024)

• Sparse Diffusion Policy: A Sparse, Reusable, and Flexible Policy for Robot Learning (CoRL 2024)

• Efficient Diffusion Transformer Policies with Mixture of Expert Denoisers for Multitask Learning (Dec 2024)

Is Conditional Generative Modeling all you need for Decision-Making? (Decision Diffuser)

• Waypoint-Based Imitation Learning for Robotic Manipulation (CoRL 2023)

• RoLD: Robot Latent Diffusion for Multi-task Policy Modeling ()

Provable Guarantees for Generative Behavior Cloning: Bridging Low-Level Stability and High-Level Behavior (Highly Theoretical)

Cold Diffusion on the Replay Buffer: Learning to Plan from Known Good States

SplatSim: Zero-Shot Sim2Real Transfer of RGB Manipulation Policies Using Gaussian Splatting (Sim2Real)

---

C3DM: Constrained-Context Conditional Diffusion Models for Imitation Learning (tackle spurious correlation, distraction free)

From Imitation to Refinement – Residual RL for Precise Assembly (distribution shifts)

• Adaptive Online Replanning with Diffusion Models (NeurIPS 2023)

Unpacking Failure Modes of Generative Policies: Runtime Monitoring of Consistency and Progress (out-of-distribution scenarios, detect failures)

Failure Prediction from Limited Hardware Demonstrations (failure prediction)

---

Crossway Diffusion: Improving Diffusion-based Visuomotor Policy via Self-supervised Learning

Discrete Policy: Learning Disentangled Action Space for Multi-Task Robotic Manipulation

INSTANT POLICY: IN-CONTEXT IMITATION LEARNING VIA GRAPH DIFFUSION (graph diffusion)

FlowBotHD: History-Aware Diffuser Handling Ambiguities in Articulated Objects Manipulation (ambiguity)

Diffusion Policy Attacker: Crafting Adversarial Attacks for Diffusion-based Policies (adversarial attack)

Towards Effective Utilization of Mixed-Quality Demonstrations in Robotic Manipulation via Segment-Level Selection and Optimization (demo data selection and filter)

Cutting Sequence Diffuser: Sim-to-Real Transferable Planning for Object Shaping by Grinding (grinding through grinding belt)

Implicit Contact Diffuser: Sequential Contact Reasoning with Latent Point Cloud Diffusion (contact)

Diff-Control: A Stateful Diffusion-based Policy for Imitation Learning (control net)

FREE FROM BELLMAN COMPLETENESS: TRAJECTORY STITCHING VIA MODEL-BASED RETURN-CONDITIONED SUPERVISED LEARNING

One-Shot Imitation under Mismatched Execution

SDP: Spiking Diffusion Policy for Robotic Manipulation with Learnable Channel-Wise Membrane Thresholds

Locomotion

• Generalizable Humanoid Manipulation with Improved 3D Diffusion Policies (Oct 2024)

• Robot Motion Diffusion Model: Motion Generation for Robotic Characters

• Diffusion-based learning of contact plans for agile locomotion

• DIDI: Diffusion-Guided Diversity for Offline Behavioral Generation (To be checked)

• Preference Aligned Diffusion Planner for Quadrupedal Locomotion Control (out-of-distribution issue)

• The Role of Domain Randomization in Training Diffusion Policies for Whole-Body Humanoid Control

Dexterous Manipulation

Navigation

Task and Motion Planning

Others

DroneDiffusion: Robust Quadrotor Dynamics Learning with Diffusion Models (Drones)

Robot Learning Utilizing Diffusion Model Properties

PoCo: Policy Composition from and for Heterogeneous Robot Learning (To be checked)

Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion (To be checked)

Pre-trained Text-to-Image Diffusion Models Are Versatile Representation Learners for Control

Imitation Learning from Purified Demonstrations (using forward & reverse diffusion process to purify imperfect demonstrations)

Human-Agent Joint Learning for Efficient Robot Manipulation Skill Acquisition (a diffusion-model-based assistive agent to learn how to assist humans in collecting data in a shared control manner)

P3-PO: Prescriptive Point Priors for Visuo-Spatial Generalization of Robot Policies (using DIFT for point correspondence)

• Adaptive Online Replanning with Diffusion Models (NeurIPS 2023)

• Diffusion reward: Learning rewards via conditional video diffusion (ECCV 2024)