mlsanityData to 7 Oct 2026

VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model

arXiv 2509.09372 ↗First posted 11 Sep 2025cs.ROcited by 140

On the map: Multimodal LLMs / vision-language models → Embodied and 3D vision-language models

also in 1 more
Abstract (arXiv)

Vision-Language-Action (VLA) models typically bridge the gap between perceptual and action spaces by pre-training a large-scale Vision-Language Model (VLM) on robotic data. While this approach greatly enhances performance, it also incurs significant training costs. In this paper, we investigate how to effectively bridge vision-language (VL) representations to action (A). We introduce VLA-Adapter, a novel paradigm designed to reduce the reliance of VLA models on large-scale VLMs and extensive pre-training. To this end, we first systematically analyze the effectiveness of various VL conditions and present key findings on which conditions are essential for bridging perception and action spaces. Based on these insights, we propose a lightweight Policy module with Bridge Attention, which autonomously injects the optimal condition into the action space. In this way, our method achieves high performance using only a 0.5B-parameter backbone, without any robotic data pre-training. Extensive experiments on both simulated and real-world robotic benchmarks demonstrate that VLA-Adapter not only achieves state-of-the-art level performance, but also offers the fast inference speed reported to date. Furthermore, thanks to the proposed advanced bridging paradigm, VLA-Adapter enables the training of a powerful VLA model in just 8 hours on a single consumer-grade GPU, greatly lowering the barrier to deploying the VLA model. Project page: https://vla-adapter.github.io/.

Key facts. Cited by 140 papers in the map (data through 7 Oct 2026). Placed in Embodied and 3D vision-language models, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 66 of its 74 references that are in the map. First posted on arXiv 11 Sep 2025.

Builds on

66 of 74 references.

Closest ancestors · 9

Same direction, newest first

  1. 2025SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics2 Jun
  2. 2025CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models27 Mar
  3. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success27 Feb
  4. 2024QUART-Online: Latency-Free Large Multimodal Language Model for Quadruped Robot Learning20 Dec
  5. 2024What Matters in Building Vision-Language-Action Models for Generalist Robots18 Dec
  6. 2024DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution4 Nov
  7. 2024HiRT: Enhancing Robotic Control with Hierarchical Robot Transformers12 Sep
  8. 2023Vision-Language Foundation Models as Effective Robot Imitators2 Nov
  9. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control28 Jul

Other references · 57

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  3. 2023Visual Instruction Tuningcited by 7,792
  4. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  5. 2022Scalable Diffusion Models with Transformerscited by 5,931
  6. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,048
  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. 2024Qwen2 Technical Reportcited by 2,076
  11. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  12. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,653
38 more
  1. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,490
  2. 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
  3. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  4. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,167
  5. 2023LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attentioncited by 776
  6. 2024RDT-1B: a Diffusion Foundation Model for Bimanual Manipulationcited by 670
  7. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 643
  8. 2021CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Taskscited by 564
  9. 2025SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Modelcited by 502
  10. 2025UniVLA: Learning to Act Anywhere with Task-centric Latent Actionscited by 386
  11. 2025WorldVLA: Towards Autoregressive Action World Modelcited by 351
  12. 2024Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Modelscited by 336
  13. 2024TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policiescited by 298
  14. 2024GR-2: A Generative Video-Language-Action Model with Web-Scale Knowledge for Robot Manipulationcited by 290
  15. 2024Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representationscited by 279
  16. 2023Zero-Shot Robotic Manipulation with Pretrained Image-Editing Diffusion Modelscited by 264
  17. 2025Transfer between Modalities with MetaQueriescited by 231
  18. 2024PaliGemma 2: A Family of Versatile VLMs for Transfercited by 204
  19. 2025DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledgecited by 203
  20. 2025Unified Video Action Modelcited by 195
  21. 2025MolmoAct: Action Reasoning Models that can Reason in Spacecited by 189
  22. 2025MemoryVLA: Perceptual-Cognitive Memory in Vision-Language-Action Models for Robotic Manipulationcited by 183
  23. 2024Predictive Inverse Dynamics Models are Scalable Learners for Robotic Manipulationcited by 159
  24. 2025ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planningcited by 152
  25. 2025Unified Vision-Language-Action Modelcited by 128
  26. 2025GraspVLA: a Grasping Foundation Model Pre-trained on Billion-scale Synthetic Action Datacited by 120
  27. 2025NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Taskscited by 119
  28. 2025GR-3 Technical Reportcited by 110
  29. 2025Being-H0: Vision-Language-Action Pretraining from Large-Scale Human Videoscited by 94
  30. 2025Galaxea Open-World Dataset and G0 Dual-System VLA Modelcited by 90
  31. 2025PD-VLA: Accelerating Vision-Language-Action Model Integrated with Action Chunking via Parallel Decodingcited by 88
  32. 2025ReconVLA: Reconstructive Vision-Language-Action Model as Effective Robot Perceivercited by 80
  33. 2025OpenHelix: A Short Survey, Empirical Analysis, and Open-Source Dual-System VLA Model for Robotic Manipulationcited by 79
  34. 2024Towards Synergistic, Generalized, and Efficient Dual-System for Robotic Manipulationcited by 71
  35. 2025Eagle 2: Building Post-Training Data Strategies from Scratch for Frontier Vision-Language Modelscited by 69
  36. 2025Hume: Introducing System-2 Thinking in Visual-Language-Action Modelcited by 56
  37. 2023QUAR-VLA: Vision-Language-Action Model for Quadruped Robotscited by 55
  38. 20254D-VLA: Spatiotemporal Vision-Language-Action Pretraining with Cross-Scene Calibrationcited by 55

50 most cited of 57 All 57 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 20; 2026: 120.

140 papers.

Sort:

By area:

  1. 2026GigaWorld-Policy: An Efficient Action-Centered World--Action Modelcited by 129
  2. 2025Spatial Forcing: Implicit Spatial Representation Alignment for Vision-language-action Modelcited by 125
  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. 2025RynnVLA-002: A Unified Vision-Language-Action and World Modelcited by 70
  6. 2025VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards in World Simulatorscited by 65
  7. 2026Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodimentscited by 42
  8. 2025Unified Diffusion VLA: Vision-Language-Action Model via Joint Discrete Denoising Diffusion Processcited by 37
  9. 2025Evo-1: Lightweight Vision-Language-Action Model with Preserved Semantic Alignmentcited by 35
  10. 2025A Survey on Efficient Vision-Language-Action Modelscited by 34
  11. 2025Don't Blind Your VLA: Aligning Visual Representations for OOD Generalizationcited by 34
  12. 2026DynamicVLA: A Vision-Language-Action Model for Dynamic Object Manipulationcited by 33
38 more
  1. 2026ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?cited by 33
  2. 2025VLSA: Vision-Language-Action Models with Plug-and-Play Safety Constraint Layercited by 32
  3. 2025HiF-VLA: Hindsight, Insight and Foresight through Motion Representation for Vision-Language-Action Modelscited by 30
  4. 2026ReMem-VLA: Empowering Vision-Language-Action Model with Memory via Dual-Level Recurrent Queriescited by 28
  5. 2026StarVLA-α: Reducing Complexity in Vision-Language-Action Systemscited by 27
  6. 2026Adaptive Action Chunking at Inference-time for Vision-Language-Action Modelscited by 24
  7. 2025AVA-VLA: Improving Vision-Language-Action models with Active Visual Attentioncited by 22
  8. 2025Mixture of Horizons in Action Chunkingcited by 20
  9. 2026Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planningcited by 20
  10. 2026FASTER: Rethinking Real-Time Flow VLAscited by 20
  11. 2026Global Prior Meets Local Consistency: Dual-Memory Augmented Vision-Language-Action Model for Efficient Robotic Manipulationcited by 19
  12. 2026Light-WAM: Efficient World Action Models with State-Fusion Action Decodingcited by 19
  13. 2026Towards Generalizable Robotic Manipulation in Dynamic Environmentscited by 17
  14. 2025AsyncVLA: Asynchronous Flow Matching for Vision-Language-Action Modelscited by 16
  15. 2025MergeVLA: Cross-Skill Model Merging Toward a Generalist Vision-Language-Action Agentcited by 15
  16. 2026On-the-Fly VLA Adaptation via Test-Time Reinforcement Learningcited by 14
  17. 2026VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizoncited by 14
  18. 2026World-Value-Action Model: Implicit Planning for Vision-Language-Action Systemscited by 13
  19. 2026Evo-Depth: A Lightweight Depth-Enhanced Vision-Language-Action Modelcited by 13
  20. 2025Mantis: A Versatile Vision-Language-Action Model with Disentangled Visual Foresightcited by 12
  21. 2026VLANeXt: Recipes for Building Strong VLA Modelscited by 12
  22. 2025DualVLA: Building a Generalizable Embodied Agent via Partial Decoupling of Reasoning and Actioncited by 11
  23. 2026Scaling Behavior Cloning Improves Causal Reasoning: An Open Model for Real-Time Video Game Playingcited by 10
  24. 2026SimVLA: A Simple VLA Baseline for Robotic Manipulationcited by 10
  25. 2026MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generationcited by 10
  26. 2026HEX: Humanoid-Aligned Experts for Cross-Embodiment Whole-Body Manipulationcited by 10
  27. 2026Rethinking the Practicality of Vision-language-action Model: A Comprehensive Benchmark and An Improved Baselinecited by 9
  28. 2026TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAMcited by 9
  29. 2026DySL-VLA: Efficient Vision-Language-Action Model Inference via Dynamic-Static Layer-Skipping for Robot Manipulationcited by 8
  30. 2026CORAL: Scalable Multi-Task Robot Learning via LoRA Expertscited by 8
  31. 2026ProgressVLA: Progress-Guided Diffusion Policy for Vision-Language Robotic Manipulationcited by 8
  32. 2026LIBERO-Para: A Diagnostic Benchmark and Metrics for Paraphrase Robustness in VLA Modelscited by 8
  33. 2026SafeVLA-Bench: A Benchmark for the Success-Safety Gap in Vision-Language-Action Modelscited by 8
  34. 2026Shallow-π: Knowledge Distillation for Flow-based VLAscited by 7
  35. 2026GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specializationcited by 7
  36. 2026RotVLA: Rotational Latent Action for Vision-Language-Action Modelcited by 7
  37. 2026HazardArena: Evaluating Semantic Safety in Vision-Language-Action Modelscited by 6
  38. 2026PokeVLA: Empowering Pocket-Sized Vision-Language-Action Model with Comprehensive World Knowledge Guidancecited by 6

50 most cited of 140 All 140 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2024TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation
  2. 2025VLA-0: Building State-of-the-Art VLAs with Zero Modification
  3. 2025Vision-Language-Action Models for Robotics: A Review Towards Real-World Applications
  4. 2025Vision Language Action Models in Robotic Manipulation: A Systematic Review
  5. 2025Vision-Language-Action (VLA) Models: Concepts, Progress, Applications and Challenges
  6. 2025Pure Vision Language Action (VLA) Models: A Comprehensive Survey
  7. 2025VITA-VLA: Efficiently Teaching Vision-Language Models to Act via Action Expert Distillation
  8. 2026CoRE-VLA: Towards Scalable and Robust Vision-Language-Action Modeling via Conditional Routing of Experts