awesome-dLLM-resources

Frequently updated list of dLLM (Diffusion Large Language Models) papers, models, and other resources.

Maintained By Suhas Pai and Xiaojun Ren. For an overview of this field, see our ICLR 2026 blog post.

Make a PR if you want to include your paper on this list or if you've found a cool paper that you think should be included in this list :)

Libraries

• dlmserve: First OSS Serving Engine for Diffusion Language Models - May 25, 2026 GitHub
• DiRL: An Efficient Training Framework for Diffusion Language Models - November 18, 2025 GitHub
• dLLM: Simple Diffusion Language Modeling - October 11, 2025 GitHub

Models

• DiffusionGemma — July 31, 2026 arXiv
• Continuous Latent Diffusion Language Model — May 7, 2026 arXiv
• W1-4B-dLLM: Whaletech Diffusion LLM — April 13, 2026 Hugging Face Demo
• LLaDA2.0: Scaling Up Diffusion Language Models to 100B — December 10, 2025 arXiv
• Causal Diffusion Language Models - December 4, 2025 Blog Post
• Soft-Masked Diffusion Language Models — October 20, 2025 arXiv
• UltraLLaDA: Scaling the Context Length to 128K for Diffusion Large Language Models — October 12, 2025 arXiv
• RND1: Simple, Scalable AR-to-Diffusion Conversion — October 9, 2025 Blog Post
• SDAR: A Synergistic Diffusion-AutoRegression Paradigm for Scalable Sequence Generation - October 7, 2025 arXiv
• Syntax-Guided Diffusion Language Models with User-Integrated Personalization — October 2, 2025 arXiv
• CoDA: Coding LM via Diffusion Adaptation — September 27, 2025 arXiv
• LLaDA-MoE: A Sparse MoE Diffusion Language Model — September 25, 2025 arXiv
• Dream 7B: Diffusion Large Language Models - August 21, 2025 arXiv
• DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation — June 25, 2025 arXiv
• Mercury: Ultra-Fast Language Models Based on Diffusion — June 17, 2025 arXiv
• LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Models — May 25, 2025 arXiv
• Large Language Diffusion Models — February 14, 2025 arXiv

Scaling Laws

• Continuous Diffusion Scales Competitively with Discrete Diffusion for Language — May 18, 2026 arXiv
• Diffusion Language Models Are Super Data Learners — October 1, 2025 GitHub
• Diffusion Beats Autoregressive in Data-Constrained Settings — July 21, 2025 arXiv

Architecture & Learning Objectives

• DSL-LLaDA: Scaling Continuous Denoising to 8B Masked Diffusion LMs - May 31, 2026 arXiv
• WaveFilter: Enhancing the Long-Context Capability of Diffusion LLMs via Wavelet-Guided KV Cache Filtering - May 30, 2026 arXiv
• Triplet-Block Diffusion RWKV - May 25, 2026 arXiv
• DLLM-JEPA: Joint Embedding Predictive Architectures for Masked Diffusion Language Models - May 24, 2026 arXiv
• DiLaDiff: Distilled Latent-Augmented Diffusion for Language Modeling - May 22, 2026 arXiv
• Drifting Objectives for Refining Discrete Diffusion Language Models - May 19, 2026 arXiv
• Dynamic Chunking for Diffusion Language Models - May 15, 2026 arXiv
• Where Should Diffusion Enter a Language Model? Geometry-Guided Hidden-State Replacement - May 14, 2026 arXiv
• BitLM: Unlocking Multi-Token Language Generation with Bitwise Continuous Diffusion - May 12, 2026 arXiv
• ELF: Embedded Language Flows - May 11, 2026 arXiv
• How to Train Your Latent Diffusion Language Model Jointly With the Latent Space - May 8, 2026 arXiv
• TextLDM: Language Modeling with Continuous Latent Diffusion - May 8, 2026 arXiv
• Token Time Continuous Diffusion for Language Modeling - May 7, 2026 arXiv
• Towards Closing the Autoregressive Gap in Language Modeling via Entropy-Gated Continuous Bitstream Diffusion - May 7, 2026 arXiv
• From AR to Diffusion: Efficiently Adapting Large Language Models with Strictly Causal and Elastic Horizons - April 11, 2026 arXiv
• Beyond Hard Masks: Progressive Token Evolution for Diffusion Language Models — January 16, 2026 arXiv, Code
• CANDI: Hybrid Discrete-Continuous Diffusion Models - October 26, 2025 arXiv
• Any-Order GPT as Masked Diffusion Model: Decoupling Formulation and Architecture — June 24, 2025 arXiv

Finetuning/RL

• Mask-Aware Policy Gradients for Diffusion Language Models — July 16, 2026 arXiv
• Read the Trace, Steer the Path: Trajectory-Aware Reinforcement Learning for Diffusion Language Models — June 3, 2026 arXiv
• Knowledge Editing in Masked Diffusion Language Models — June 2, 2026 arXiv
• Revise, Don't Freeze: Sampler-Matched Training for Self-Correcting Masked Diffusion Language Models — May 31, 2026 arXiv
• Learnability-Informed Fine-Tuning of Diffusion Language Models — May 21, 2026 arXiv
• Beyond Mode-Seeking RL: Trajectory-Balance Post-Training for Diffusion Language Models — May 13, 2026 arXiv
• Self-Distilled Trajectory-Aware Boltzmann Modeling: Bridging the Training-Inference Discrepancy in Diffusion Language Models — May 12, 2026 arXiv
• Relative Score Policy Optimization for Diffusion Language Models — May 11, 2026 arXiv
• DACA-GRPO: Denoising-Aware Credit Assignment for Reinforcement Learning in Diffusion Language Models — May 8, 2026 arXiv
• Guidance Is Not a Hyperparameter: Learning Dynamic Control in Diffusion Language Models — May 8, 2026 arXiv
• FlowLM: Few-Step Language Modeling via Diffusion-to-Flow Adaptation — April 6, 2026 arXiv
• Principled RL for Diffusion LLMs Emerges from a Sequence-Level Perspective — December 3, 2025 arXiv
• Diffusion LLM with Native Variable Generation Lengths: Let [EOS] Lead the Way — October 28, 2025 arXiv
• Principled and Tractable RL for Reasoning with Diffusion Language Models — October 5, 2025 arXiv
• Fine-Tuning Masked Diffusion for Provable Self-Correction — October 1, 2025 arXiv
• WeFT: Weighted Entropy-Driven Fine-Tuning for dLLMs — September 25, 2025 arXiv
• Inpainting-Guided Policy Optimization for Diffusion Large Language Models — September 12, 2025 arXiv
• Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models — September 8, 2025 arXiv

Distillation

• SelFusion: Self-distillation for Diffusion Language Models — August 24, 2026 arXiv
• Learning from the Self-future: On-policy Self-distillation for dLLMs — June 16, 2026 arXiv, Code
• Infinite Mask Diffusion for Few-Step Distillation — May 11, 2026 arXiv
• TAD: Temporal-Aware Trajectory Self-Distillation for Fast and Accurate Diffusion LLM — May 10, 2026 arXiv

Reasoning

• TiDAR: Think in Diffusion, Talk in Autoregression — November 12, 2025 arXiv
• Enhancing Reasoning for Diffusion LLMs via Distribution Matching Policy Optimization — October 9, 2025 arXiv
• Improving Reasoning for Diffusion Language Models via Group Diffusion Policy Optimization — October 9, 2025 arXiv
• DiFFPO: Training Diffusion LLMs to Reason Fast and Furious via Reinforcement Learning — October 2, 2025 arXiv
• RFG: Test-Time Scaling for Diffusion Large Language Model Reasoning with Reward-Free Guidance — September 29, 2025 arXiv
• Thinking Inside the Mask: In-Place Prompting in Diffusion LLMs — August 14, 2025 arXiv
• d1: Scaling Reasoning in Diffusion Large Language Models via Reinforcement Learning — June 3, 2025 arXiv

Decoding Strategies

• Dependency-Aware Revocable Decoding for Efficient Diffusion Large Language Model Inference — August 27, 2026 arXiv
• Information-Guided Frontier Decoding: Contextual Utility-Driven Commitment in dMLLMs — August 27, 2026 arXiv
• Survival-Guided Length Control for Efficient Diffusion Language Models — August 26, 2026 arXiv
• Context-Aware Cluster Decoding: Semantic Anchor-Driven Coherence in dMLLMs — August 23, 2026 arXiv
• Length-Adaptive Decoding for Masked Diffusion Machine Translation — August 23, 2026 arXiv
• NAVIRA: Decoupled Stochastic Remasking for Masked Diffusion Language Models — June 4, 2026 arXiv
• SemBlock: Semantic Boundary Dynamic Blocks for Diffusion LLMs — June 3, 2026 arXiv
• Dynamic Infilling Anchors for Format-Constrained Generation in Diffusion Large Language Models — June 3, 2026 arXiv
• Supportive Token Revealing for Fast Diffusion Language Model Decoding — June 2, 2026 arXiv
• DLM-SWAI: Steering Diffusion Language Models Before They Unmask — May 28, 2026 arXiv
• When Confidence Misleads: Suffix Anchoring and Anchor-Proximity Confidence Modulation for Diffusion Language Models — May 27, 2026 arXiv
• The Path Matters: Learning a Token-Commitment Policy for Diffusion Language Models — May 23, 2026 arXiv
• Constrained Code Generation with Discrete Diffusion — May 16, 2026 arXiv
• Roll Out and Roll Back: Diffusion LLMs are Their Own Efficiency Teachers — May 16, 2026 arXiv
• Adaptive Steering and Remasking for Safe Generation in Diffusion Language Models — May 13, 2026 arXiv
• Edit-Based Refinement for Parallel Masked Diffusion Language Models — May 10, 2026 arXiv
• Steering Without Breaking: Mechanistically Informed Interventions for Discrete Diffusion Language Models — May 8, 2026 arXiv
• When to Commit? Towards Variable-Size Self-Contained Blocks for Discrete Diffusion Language Models — April 27, 2026 arXiv
• Remask, Don't Replace: Token-to-Mask Refinement in Masked Diffusion Language Models — April 20, 2026 arXiv
• Stability-Weighted Decoding for Diffusion Language Models — April 18, 2026 arXiv
• Targeted Remasking: Replacing Token Editing with Token-to-Mask Refinement in Discrete Diffusion Language Models — April 6, 2026 arXiv
• LogicDiff: Logic-Guided Denoising Improves Reasoning in Masked Diffusion Language Models — March 24, 2026 arXiv
• GeoBlock: Inferring Block Granularity from Dependency Geometry in Diffusion Language Models — March 4, 2026 arXiv
• Efficient Self-Evaluation for Diffusion Language Models via Sequence Regeneration — March 3, 2026 arXiv
• One Token Is Enough: Improving Diffusion Language Models with a Sink Token — January 27, 2026 arXiv
• Deferred Commitment Decoding for Diffusion Language Models with Confidence-Aware Sliding Windows — January 5, 2026 arXiv
• Activation Steering for Masked Diffusion Language Models — December 30, 2025 arXiv
• ReFusion: A Diffusion Large Language Model with Parallel Autoregressive Decoding — December 15, 2025 arXiv
• Beyond Static Cutoffs: One-Shot Dynamic Thresholding for Diffusion Language Models — November 3, 2025 arXiv
• CreditDecoding: Accelerating Parallel Decoding in Diffusion Large Language Models with Trace Credits — October 7, 2025 arXiv
• AdaBlock-dLLM: Semantic-Aware Diffusion LLM Inference via Adaptive Block Size — September 30, 2025 arXiv
• Don't Settle Too Early: Self-Reflective Remasking for Diffusion Language Models — September 28, 2025 arXiv
• Time Is a Feature: Exploiting Temporal Dynamics in Diffusion Language Models — August 12, 2025 arXiv
• Beyond Fixed: Training-Free Variable-Length Denoising for Diffusion Large Language Models — August 1, 2025 arXiv
• Unveiling the Potential of Diffusion Large Language Model in Controllable Generation — July 6, 2025 arXiv
• Remasking Discrete Diffusion Models with Inference-Time Scaling — March 1, 2025 arXiv

Speculative Decoding

• SimSD: Simple Speculative Decoding in Diffusion Language Models — June 1, 2026 arXiv
• DFlare: Scaling Up Draft Capacity for Block Diffusion Speculative Decoding — June 1, 2026 arXiv
• Cost-Aware Diffusion Draft Trees for Speculative Decoding — June 1, 2026 arXiv
• PSD: Pushing the Pareto Frontier of Diffusion LLMs via Parallel Speculative Decoding — May 15, 2026 arXiv
• Factorization-Error-Free Discrete Diffusion Language Model via Speculative Decoding — May 14, 2026 arXiv
• Self-Speculative Decoding for Diffusion Large Language Models — October 5, 2025 arXiv
• Spiffy: Multiplying Diffusion LLM Acceleration via Lossless Speculative Decoding — September 22, 2025 arXiv

Inference Optimization

• Accelerating Diffusion Language Models via Structured Suffix Modeling — August 24, 2026 arXiv
• SAID: Accelerating Diffusion-Based Language Models via Scaffold-Aware Iterative Decoding — June 3, 2026 arXiv
• Fast-dLLM++: Fréchet Profile Decoding for Faster Diffusion LLM Inference — June 1, 2026 arXiv, Code, Project Page
• EPIC: Efficient and Parallel Inference under CFG Constraints for Diffusion Language Models — May 30, 2026 arXiv
• Efficient Diffusion LLMs via Temporal-Spatial Parallel Decoding and Confidence Extrapolation — May 29, 2026 arXiv
• PulseCol: Periodically Refreshed Column-Sparse Attention for Accelerating Diffusion Language Models — May 20, 2026 arXiv
• TIDE: Efficient and Lossless MoE Diffusion LLM Inference with I/O-aware Expert Offload — May 19, 2026 arXiv
• Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMs — May 7, 2026 arXiv
• R2-dLLM: Accelerating Diffusion Large Language Models via Spatio-Temporal Redundancy Reduction — April 21, 2026 arXiv
• Streaming-dLLM: Accelerating Diffusion LLMs via Suffix Pruning and Dynamic Decoding — January 25, 2026 arXiv
• CD4LM: Consistency Distillation and aDaptive Decoding for Diffusion Language Models — January 5, 2026 arXiv
• WeDLM: Reconciling Diffusion Language Models with Standard Causal Attention for Fast Inference — December 28, 2025 arXiv
• Mask Tokens as Prophet: Fine-Grained Cache Eviction for Efficient dLLM Inference — October 10, 2025 arXiv
• dInfer: An Efficient Inference Framework for Diffusion Language Models — October 9, 2025 arXiv
• Accelerating Diffusion LLM Inference via Local Determinism Propagation — October 8, 2025 arXiv
• Free Draft-and-Verification: Toward Lossless Parallel Decoding for Diffusion Large Language Models — September 30, 2025 arXiv
• Fast-dLLM v2: Efficient Block-Diffusion LLM — September 30, 2025 arXiv
• Learning to Parallel: Accelerating Diffusion Large Language Models via Learnable Parallel Decoding — September 29, 2025 (ICLR 2026) arXiv
• Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct — September 29, 2025 arXiv
• d²Cache: Accelerating Diffusion-Based LLMs via Dual Adaptive Caching — September 27, 2025 arXiv
• Sparse-dLLM: Accelerating Diffusion LLMs with Dynamic Cache Eviction — August 4, 2025 arXiv
• Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principles — June 12, 2025 arXiv
• Fast-dLLM: Training-Free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding — May 28, 2025 arXiv
• Accelerating Diffusion LLMs via Adaptive Parallel Decoding — May 31, 2025 arXiv
• dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching — May 17, 2025 arXiv

Quantization

• On the Quantization Robustness of Diffusion Language Models in Coding Benchmarks — April 22, 2026 arXiv
• Quant-dLLM: Post-Training Extreme Low-Bit Quantization for Diffusion Large Language Models — September 27, 2025 arXiv
• Quantization Meets dLLMs: A Systematic Study of Post-Training Quantization for Diffusion LLMs — August 20, 2025 arXiv
• Dllmquant: Quantizing Diffusion-Based Large Language Models — August 14, 2025 arXiv

Retrieval

• Self-Augmenting Retrieval for Diffusion Language Models — June 4, 2026 arXiv
• DiffRetriever: Parallel Representative Tokens for Retrieval with Diffusion Language Models — May 8, 2026 arXiv

Security & Privacy

• SAC-Copula: Quality-Preserving Watermarking for Diffusion Language Models via Smooth Correlated Gumbel Fields — August 21, 2026 arXiv
• Global Sketch-Based Watermarking for Diffusion Language Models — June 3, 2026 arXiv
• Extracting Training Data from Diffusion Language Models via Infilling — May 22, 2026 arXiv
• Machine Unlearning for Masked Diffusion Language Models — May 18, 2026 arXiv
• Chainwash: Multi-Step Rewriting Attacks on Diffusion Language Model Watermarks — May 6, 2026 arXiv
• TrustLDM: Benchmarking Trustworthiness in Language Diffusion Models — April 15, 2026 arXiv
• GCG Attack On A Diffusion LLM — December 30, 2025 arXiv
• DiffuGuard: How Intrinsic Safety Is Lost and Found in Diffusion Large Language Models — September 29, 2025 arXiv
• Where to Start Alignment? Diffusion Large Language Model May Demand a Distinct Position — August 17, 2025 arXiv
• Jailbreaking Large Language Diffusion Models: Revealing Hidden Safety Flaws in Diffusion-Based Text Generation — July 25, 2025 arXiv
• The Devil Behind the Mask: An Emergent Safety Vulnerability of Diffusion LLMs — July 15, 2025 arXiv

Theory

• Language Generation as Optimal Control: Closed-Loop Diffusion in Latent Control Space — May 14, 2026 arXiv
• Optimal inference schedules for masked diffusion models — November 9, 2025 arXiv
• On Powerful Ways to Generate: Autoregression, Diffusion, and Beyond — October 7, 2025 arXiv
• Why mask diffusion does not work — September 25, 2025 arXiv

Applications

• DiG-Plan: Mitigating Early Commitment for Tool-Graph Planning via Diffusion Guidance — June 4, 2026 arXiv
• DiffuSent: Towards a Unified Diffusion Framework for Aspect-Based Sentiment Analysis — May 31, 2026 arXiv
• DiffScore: Text Evaluation Beyond Autoregressive Likelihood — May 12, 2026 arXiv
• BiMol-Diff: A Unified Diffusion Framework for Molecular Generation and Captioning — April 27, 2026 arXiv
• Harnessing LLM for Noise-Robust Cognitive Diagnosis in Web-Based Intelligent Education Systems — October 5, 2025 arXiv
• Discovering Mathematical Equations with Diffusion Language Model — September 16, 2025 arXiv

Analysis/Position Papers

• What Gets Unmasked First? Trajectory Analysis of Diffusion Models for Graph-to-Text Generation — May 29, 2026 arXiv
• The Confidence Shortcut: A Reasoning Failure Mode of Masked Diffusion Models — May 27, 2026 arXiv
• Prompt Compression in Diffusion Large Language Models: Evaluating LLMLingua-2 on LLaDA — May 18, 2026 arXiv
• Uncertainty Quantification for Large Language Diffusion Models — May 14, 2026 arXiv
• Understanding and Accelerating the Training of Masked Diffusion Language Models — May 13, 2026 arXiv
• Differences in Text Generated by Diffusion and Autoregressive Language Models — April 4, 2026 arXiv
• Efficient Self-Evaluation for Diffusion Language Models via Sequence Regeneration — March 3, 2026 (ACL 2026) arXiv
• On the Role of Discreteness in Diffusion LLMs — December 27, 2025 arXiv
• Understanding the Limitations of Diffusion LLMs through a Probabilistic Perspective - November 25, 2025 Notion

Surveys

• Accelerating Masked Diffusion Large Language Models: A Survey of Efficient Inference Techniques — July 14, 2026 arXiv
• dLLM - Rethinking Generation Beyond Autoregressive Models — April 27, 2026 ICLR Blog
• Beyond Autoregression: An Empirical Study of Diffusion Large Language Models for Code Generation — September 14, 2025 arXiv
• Diffusion-Based Large Language Models Survey — August 26, 2025 TechRxiv
• A Survey on Diffusion Language Models — August 14, 2025 arXiv
• Discrete Diffusion in Large Language and Multimodal Models: A Survey — June 16, 2025 arXiv