mlsanityData to 7 Oct 2026

Generalist VLA models and benchmarks

276 papers0 this weekMedium confidence in this grouping: mostly one topic, with blurry edgesMedium confidence in this grouping: mostly one topic, with blurry edges · About

Keywords: vla, vision-language-action, robot, robotic, manipulation

Key facts. Generalist VLA models and benchmarks: 276 papers on arXiv since 2023 (data through 7 Oct 2026). 225 of its 276 mapped papers (82%) cite “OpenVLA: An Open-Source Vision-Language-Action Model” (2024, arXiv 2406.09246). 202 of its 276 mapped papers (73%) cite “RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control” (2023, arXiv 2307.15818). 14 new papers 17 Sep–7 Oct.

What’s new

17 Sep – 7 Oct 2026

14 new papers, 17 Sep – 7 Oct; 0 this week.

  1. Earlier · 14
  2. When Does Retrieval Help? A Study of In-Context Adaptation in Vision-Language-Action Models also fits: VLA architectures and generalization
  3. Vela: Scaling Vision-Language-Action Models with Adaptive Action Curve Parametrization also fits: World model-based VLA
  4. MixVLA: Adaptive Mixing of Non-Invariant Information for Generalizable Vision-Language-Action Models also fits: Human-video and latent-action pretraining
  5. DuoMind: Enabling Distributed Multi-Robot Coordination with Semantic Communication also fits: Reasoning and chain-of-thought in VLA
  6. Same Scene, Different Task: Skill Alignment for Compositional Generalization in VLAs also fits: VLA architectures and generalization
  7. MotorMind: Scaffolding General Vision Language Models for Zero-Shot Robot Manipulation also fits: Reasoning and chain-of-thought in VLA
  8. InfraVLA: Extending Vision-Language-Action Navigation with Infrastructure Cameras also fits: 3D and spatial-aware VLA
  9. TAO-DA: Towards Autonomous Operation--A Dual-Arm Vision-Language-Action Model for Coordinated Manipulation also fits: Reasoning and chain-of-thought in VLA
  10. GT-VLA: Target-Conditioned Trace Guidance for Generalizable Robotic Manipulation also fits: Reasoning and chain-of-thought in VLA
  11. MedVLA: A Hierarchical Vision-Language-Action Framework for Closed-Loop Precision Medical Robot Manipulation also fits: Reasoning and chain-of-thought in VLA
All 14
  1. MaskVLA: Visual Masking Against Trajectory Overfitting of Vision-Language-Action Model also fits: Efficient VLA inference and adaptation
  2. H-VLA: Hierarchical Vision-Language-Action Model with Key-Action Reasoning and Motion Planning in a Unified Action Space also fits: Reasoning and chain-of-thought in VLA
  3. SkipVLA: Skipping VLA Steps with Classical Planning for Fast Robot Manipulation also fits: Reinforcement learning and post-training
  4. Co-VLA: Consensus-based Federated Training for Vision-Language-Action Models also fits: Reinforcement learning and post-training

Foundations

Most cited by papers here

  1. OpenVLA: An Open-Source Vision-Language-Action Model2024cited by 225 here
  2. RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control2023cited by 202 here
  3. π0: A Vision-Language-Action Flow Model for General Robot Control2024cited by 180 here
  4. π0.5: a Vision-Language-Action Model with Open-World Generalization2025cited by 143 here
  5. Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success2025cited by 92 here

Lineage

The papers this direction builds on, oldest first.

Each paper on the line cites the one before.

  1. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 202 here
  2. 2024Octo: An Open-Source Generalist Robot Policycited by 142 here
  3. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 225 here
  4. 2024TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulationcited by 53 here
  5. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 180 here
  6. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 84 here
  7. 2025DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Controlcited by 35 here
  8. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 143 here
  9. 2025Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Bettercited by 34 here
  10. 2025X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Modelcited by 27 here
14 more it builds on, not in this chain
  1. 20243D-VLA: A 3D Vision-Language-Action Generative World Modelcited by 43 here
  2. 2024A Survey on Vision-Language-Action Models for Embodied AIcited by 33 here
  3. 2024Robotic Control via Embodied Chain-of-Thought Reasoningcited by 50 here
  4. 2024Latent Action Pretraining from Videoscited by 26 here
  5. 2024CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulationcited by 51 here
  6. 2024TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policiescited by 41 here
  7. 2025SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Modelcited by 65 here
  8. 2025ChatVLA: Unified Multimodal Understanding and Robot Control with Vision-Language-Action Modelcited by 31 here
  9. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 92 here
  10. 2025CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Modelscited by 50 here
  11. 2025UniVLA: Learning to Act Anywhere with Task-centric Latent Actionscited by 38 here
  12. 2025SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Roboticscited by 60 here
  13. 2025MemoryVLA: Perceptual-Cognitive Memory in Vision-Language-Action Models for Robotic Manipulationcited by 28 here
  14. 2025LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Modelscited by 31 here

The 24 most cited of 475.

Browse

30 of 276, ranked 6–35

  1. 6SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics202560
  2. 7TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation202453
  3. 8CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation202451
  4. 9DexVLA: Vision-Language Model with Plug-In Diffusion Expert for General Robot Control202535
  5. 10Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Better202534
  6. 11A Survey on Vision-Language-Action Models for Embodied AI202433
  7. 12X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model202527
  8. 13GR-3 Technical Report202525
  9. 14HAMSTER: Hierarchical Action Models For Open-World Robot Manipulation202524
  10. 15NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Tasks202523
All 30
  1. 16OpenHelix: A Short Survey, Empirical Analysis, and Open-Source Dual-System VLA Model for Robotic Manipulation202521
  2. 17VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model202520
  3. 18What Matters in Building Vision-Language-Action Models for Generalist Robots202420
  4. 19Unified Vision-Language-Action Model202517
  5. 20UP-VLA: A Unified Understanding and Prediction Model for Embodied Agent202516
  6. 21RoboMamba: Efficient Vision-Language-Action Model for Robotic Reasoning and Manipulation202416
  7. 22Towards Synergistic, Generalized, and Efficient Dual-System for Robotic Manipulation202416
  8. 23Actions as Language: Fine-Tuning VLMs into VLAs Without Catastrophic Forgetting202515
  9. 24HiRT: Enhancing Robotic Control with Hierarchical Robot Transformers202415
  10. 25QUAR-VLA: Vision-Language-Action Model for Quadruped Robots202315
  11. 26VLAS: Vision-Language-Action Model With Speech Instructions For Customized Robot Manipulation202514
  12. 27OTTER: A Vision-Language-Action Model with Text-Aware Visual Feature Extraction202513
  13. 28Galaxea Open-World Dataset and G0 Dual-System VLA Model202513
  14. 29ReconVLA: Reconstructive Vision-Language-Action Model as Effective Robot Perceiver202512
  15. 30FLOWER: Democratizing Generalist Robot Policies with Efficient Vision-Language-Action Flow Policies202512
  16. 31LLaRA: Supercharging Robot Learning Data for Vision-Language Policy202412
  17. 32NaVILA: Legged Robot Vision-Language-Action Model for Navigation202412
  18. 33A Pragmatic VLA Foundation Model202611
  19. 34Beyond Sight: Finetuning Generalist Robot Policies with Heterogeneous Sensors via Language Grounding202510
  20. 35Interleave-VLA: Enhancing Robot Manipulation with Interleaved Image-Text Instructions202510