mlsanityData to 7 Oct 2026

Discrete Diffusion VLA: Bringing Discrete Diffusion to Action Decoding in Vision-Language-Action Policies

arXiv 2508.20072 ↗First posted 27 Aug 2025cs.CV cs.LG cs.ROcited by 90

On the map: Vision-language-action models → Efficient VLA inference and adaptation

also in 2 more
Abstract (arXiv)

Vision-Language-Action (VLA) models adapt large vision-language backbones to map images and instructions into robot actions. However, prevailing VLAs either generate actions autoregressively in a fixed left-to-right order with poor performance or attach separate diffusion heads outside the backbone that fragments information pathways and hinders unified, scalable architectures. Instead, we present Discrete Diffusion VLA that discretizes action chunks and models them with discrete diffusion pattern retaining progressive refinement inside the unified transformer backbone. Our method achieves an adaptive decoding order that resolves high-confidence action elements before harder ones and employs secondary re-masking to revisit uncertain predictions, enabling robust error correction. This design preserves pretrained vision-language priors, supports parallel decoding, and improves the efficiency. Discrete Diffusion VLA achieves 96.4% avg. success on LIBERO, 71.2% visual matching on SimplerEnv-Fractal, and 54.2% overall on SimplerEnv-Bridge. On out-of-distribution tests of LIBERO-Goal, our method exhibits only 0.8% language degradation versus 8.0% of parallel decoding, and 20.4% vision degradation versus 29.0% for continuous diffusion, demonstrating well retention of pretrained vision-language capabilities. We also conduct two real-robot evaluations on AgileX Cobot Magic platform to show the method's effectiveness.

Key facts. Cited by 90 papers in the map (data through 7 Oct 2026). Placed in Efficient VLA inference and adaptation, one of 13 directions in Vision-language-action models. Builds on 60 of its 75 references that are in the map. First posted on arXiv 27 Aug 2025.

Builds on

60 of 75 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025Dita: Scaling Diffusion Transformer for Generalist Vision-Language-Action Policy25 Mar
  2. 2025HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Model13 Mar
  3. 2025PD-VLA: Accelerating Vision-Language-Action Model Integrated with Action Chunking via Parallel Decoding4 Mar
  4. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Models16 Jan

Other references · 56

  1. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  3. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  4. 2017Neural Discrete Representation Learningcited by 4,650
  5. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,048
  6. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  7. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  8. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
  9. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,297
  10. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  11. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,665
  12. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,653
38 more
  1. 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
  2. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  3. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
  4. 2022MaskGIT: Masked Generative Image Transformercited by 1,082
  5. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,023
  6. 2025Large Language Diffusion Modelscited by 872
  7. 2023Open X-Embodiment: Robotic Learning Datasets and RT-X Modelscited by 849
  8. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 844
  9. 2021Vector Quantized Diffusion Model for Text-to-Image Synthesiscited by 773
  10. 2024PaliGemma: A versatile 3B VLM for transfercited by 679
  11. 2024RDT-1B: a Diffusion Foundation Model for Bimanual Manipulationcited by 670
  12. 2023BridgeData V2: A Dataset for Robot Learning at Scalecited by 634
  13. 2024Simplified and Generalized Masked Diffusion for Discrete Datacited by 606
  14. 2025SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Roboticscited by 507
  15. 2025SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Modelcited by 502
  16. 2024Evaluating Real-World Robot Manipulation Policies in Simulationcited by 484
  17. 2024Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Modelcited by 468
  18. 2024CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulationcited by 411
  19. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decodingcited by 404
  20. 2025AgiBot World Colosseo: A Large-scale Manipulation Platform for Scalable and Intelligent Embodied Systemscited by 390
  21. 2025UniVLA: Learning to Act Anywhere with Task-centric Latent Actionscited by 386
  22. 2024Better & Faster Large Language Models via Multi-token Predictioncited by 344
  23. 2024Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Modelscited by 336
  24. 2024Latent Action Pretraining from Videoscited by 302
  25. 2024Discrete Flow Matchingcited by 301
  26. 2024TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policiescited by 298
  27. 2024Scaling Diffusion Language Models via Adaptation from Autoregressive Modelscited by 268
  28. 2025MMaDA: Multimodal Large Diffusion Language Modelscited by 241
  29. 2022DiffusionBERT: Improving Generative Masked Language Models with Diffusion Modelscited by 214
  30. 2025Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusionscited by 208
  31. 2024Masked Diffusion Models are Secretly Time-Agnostic Masked Models and Exploit Inaccurate Categorical Samplingcited by 204
  32. 2025DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Controlcited by 191
  33. 2024Predictive Inverse Dynamics Models are Scalable Learners for Robotic Manipulationcited by 159
  34. 2023AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Plannerscited by 147
  35. 2024Scaling Proprioceptive-Visual Learning with Heterogeneous Pre-trained Transformerscited by 145
  36. 2024Multimodal Diffusion Transformer: Learning Versatile Behavior from Multimodal Goalscited by 131
  37. 2025DexGraspVLA: A Vision-Language-Action Framework Towards General Dexterous Graspingcited by 87
  38. 2025InstructVLA: Vision-Language-Action Instruction Tuning from Understanding to Manipulationcited by 71

50 most cited of 56 All 56 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 22; 2026: 68.

90 papers.

Sort:

By area:

  1. 2026Causal World Modeling for Robot Controlcited by 289
  2. 2025X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Modelcited by 285
  3. 2026VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Modelcited by 104
  4. 2026ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learningcited by 94
  5. 2026Being-H0.7: A Latent World-Action Model from Egocentric Videoscited by 68
  6. 2026Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalizationcited by 61
  7. 2026Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Modelscited by 47
  8. 2025Pure Vision Language Action (VLA) Models: A Comprehensive Surveycited by 46
  9. 2026Xiaomi-Robotics-0: An Open-Sourced Vision-Language-Action Model with Real-Time Executioncited by 42
  10. 2026AHA-WAM:Asynchronous Horizon-Adaptive World-Action Modeling with Observation-Guided Context Routingcited by 41
  11. 2026RMBench: Memory-Dependent Robotic Manipulation Benchmark with Insights into Policy Designcited by 40
  12. 2025Unified Diffusion VLA: Vision-Language-Action Model via Joint Discrete Denoising Diffusion Processcited by 37
38 more
  1. 2025A Survey on Efficient Vision-Language-Action Modelscited by 34
  2. 2025LLaDA-VLA: Vision Language Diffusion Action Modelscited by 30
  3. 2025dVLA: Diffusion Vision-Language-Action Model with Multimodal Chain-of-Thoughtcited by 30
  4. 2026ACoT-VLA: Action Chain-of-Thought for Vision-Language-Action Modelscited by 30
  5. 2025VLA-Arena: An Open-Source Framework for Benchmarking Vision-Language-Action Modelscited by 28
  6. 2026ABot-M0.5: Unified Mobility-and-Manipulation World Action Modelcited by 28
  7. 2025SwiftVLA: Unlocking Spatiotemporal Dynamics for Lightweight VLA Models at Minimal Overheadcited by 23
  8. 2025Dream-VL & Dream-VLA: Open Vision-Language and Vision-Language-Action Models with Diffusion Language Model Backbonecited by 18
  9. 2026Learning Native Continuation for Action Chunking Flow Policiescited by 18
  10. 2026EventVLA: Event-Driven Visual Evidence Memory for Long-Horizon Vision-Language-Action Policiescited by 18
  11. 2025Exploring the Potential of Diffusion Large Language Models in Code Generationcited by 16
  12. 2025AsyncVLA: Asynchronous Flow Matching for Vision-Language-Action Modelscited by 16
  13. 2025dVLM-AD: Enhance Diffusion Vision-Language-Model for Driving via Controllable Reasoningcited by 14
  14. 2026Learning 4D Geometric Priors for Inference-Efficient World Action Modelscited by 14
  15. 2026World-Value-Action Model: Implicit Planning for Vision-Language-Action Systemscited by 13
  16. 2025Expertise need not monopolize: Action-Specialized Mixture of Experts for Vision-Language-Action Learningcited by 12
  17. 2025Unifying Perception and Action: A Hybrid-Modality Pipeline with Implicit Visual Chain-of-Thought for Robotic Action Generationcited by 12
  18. 2026dWorldEval: Scalable Robotic Policy Evaluation via Discrete Diffusion World Modelcited by 11
  19. 2026SimVLA: A Simple VLA Baseline for Robotic Manipulationcited by 10
  20. 2026MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generationcited by 10
  21. 2026Fast-dVLA: Accelerating Discrete Diffusion VLA to Real-Time Performancecited by 9
  22. 2026TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAMcited by 9
  23. 2026LLaDA-o: An Effective and Length-Adaptive Omni Diffusion Modelcited by 8
  24. 2026CORAL: Scalable Multi-Task Robot Learning via LoRA Expertscited by 8
  25. 2025LatBot: Distilling Universal Latent Actions for Vision-Language-Action Modelscited by 7
  26. 2025Unified Embodied VLM Reasoning with Robotic Action via Autoregressive Discretized Pre-trainingcited by 7
  27. 2026Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularizationcited by 7
  28. 2026PoseVLA: Universal Pose Pretraining for Generalizable Vision-Language-Action Policiescited by 7
  29. 2026GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specializationcited by 7
  30. 2026Green-VLA: Staged Vision-Language-Action Model for Generalist Robotscited by 6
  31. 2026VPWEM: Non-Markovian Visuomotor Policy with Working and Episodic Memorycited by 6
  32. 2026Libra-VLA: Achieving Learning Equilibrium via Asynchronous Coarse-to-Fine Dual-Systemcited by 6
  33. 2026PRTS: A Primitive Reasoning and Tasking System via Contrastive Representationscited by 6
  34. 2026PriorVLA: Prior-Preserving Adaptation for Vision-Language-Action Modelscited by 6
  35. 2025Co-GRPO: Co-Optimized Group Relative Policy Optimization for Masked Diffusion Modelcited by 5
  36. 2026DFM-VLA: Iterative Action Refinement for Robot Manipulation via Discrete Flow Matchingcited by 5
  37. 2026Rethinking Muon Beyond Pretraining: Spectral Failures and High-Pass Remedies for VLA and RLVRcited by 5
  38. 2025An Anatomy of Vision-Language-Action Models: From Modules to Milestones and Challengescited by 4

50 most cited of 90 All 90 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Reducing Temporal Redundancy for Efficient Vision-Language-Action Inference
  2. 2026StairVLA: Stage-Aware Hierarchical Action Generation for Vision-Language-Action Models
  3. 2025EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models
  4. 2025VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model
  5. 2026CHASE-VLA: Post-Training Quantization Framework for Vision-Language-Action Models with Chunk-Aware Scale Estimation
  6. 2025Robotic VLA Benefits from Joint Learning with Motion Image Diffusion
  7. 2026Action Draft and Verify: A Self-Verifying Framework for Vision-Language-Action Model
  8. 2026A1: A Fully Transparent Open-Source, Adaptive and Efficient Truncated Vision-Language-Action Model