Robustness of VLAs

VLA: Vision-Language-Action Model

Overview

This field studies whether robots guided by vision-language-action models can follow instructions and act reliably when viewpoints, lighting, objects, sensor inputs, or robot execution differ from training. Training approaches include perturbation and counterfactual data, consistency objectives, and world-model supervision; inference-time methods use adaptation, monitoring, action correction, and failure recovery. Benchmarks such as LIBERO-Plus and RoboTwin test shifts in simulation, while real-robot evaluations expose hardware faults and contact-related failures; increasingly, evaluations measure safety and behavior over a rollout, not just task completion. A central challenge is improving performance under distribution shift without weakening clean-task performance, sensitivity to meaningful instructions, or safe behavior.

Latest papers

CardsList

239 papers20 papers in the last four weeks, down 67% on the four weeks before. 0.2% of all new papers.

  1. RECLAIM: Robot-Vision-Informed Adversarial Patch Defense via Embodied Attention Feedback

    Nov 18, 2026Jiayou Zhu, Peng Guo, Litian Zhang +6Robustness of VLAsAdversarial Patch Attacks

  2. Homeward: No Place Like the Training View

    Nov 18, 2026Mingyang Lyu, Yinqian Sun, Gongce Cao +8Robustness of VLAsVLA Adaptation

  3. Towards Pose-Robust Few-Shot Vision-Language-Action Models

    Nov 18, 2026Wenhao Li, Zhongyi Han, Yuhao Wei +3Robustness of VLAsVLA Adaptation

  4. ExecGap: Exploiting Command-Execution Gap for Patch Attacks on Vision-Language-Action Models

    Nov 18, 2026Keke Tang, Guolin Qiu, Tianyu Hao +4Adversarial Attacks on VLAsRobustness of VLAs

  5. CoBoVLA: Decoupling Language from Action with Discrete Codes for Robust and Interpretable Robot Manipulation

    Nov 17, 2026Mehul Arora, Dhruv Agarwal, Ambuj AgrawalVision-Language-Action ModelsInstruction Generalization in VLAs

  6. Right Goal, Wrong Body: Re-grounding Frozen VLAs under Actuation Shift

    Nov 15, 2026Umer Siddique, Peilang Li, Yongcan CaoTest-Time Adaptation of VLAsVLA Adaptation

  7. Self-Calibrating Vision-Agent-Action: Deploying Pick-and-Place to New Robot Cells without Demonstrations or Fine-Tuning

    Nov 14, 2026Jae Hun Hwang, Yechan Seo, Hong-In WonVision-Based Robot ControlZero-Shot Robotic Manipulation

  8. MOSAIC: Modular Obstacle and Spatial-Aware Framework for Manipulation In Cluttered Scenes

    Nov 11, 2026Richard Shi, Mohammad Mahdavian, Mehrdad Momeni Zadeh +2Spatial Grounding for VLAsVision-Language-Action Models

  9. How Much of Your VLA Is kNN

    Nov 9, 2026Denis Grachev, Damien LaRocque, Omar MouredVision-Language-Action ModelsVLA Evaluation

  10. Failure Prediction and Classification in Vision-Language-Action Models Via Calibrated Latent Probing

    Nov 5, 2026Michal P. Podolinsky, Yunhao Yang, Daniel Little +6Failure Detection for VLAsFailure Prediction for VLAs

  11. T3R: Training-Free Two-Stage Token Refinement Towards Efficient and Robust VLA Models

    Nov 5, 2026Mohammed Hassan, Zhenyang Chen, Zheng Wang +5Efficient VLA InferenceVisual Token Pruning

  12. From Pixels to Actions--Towards Native Vision-Language-Action Policies

    Nov 4, 2026Weichen Fan, Haiwen Diao, Penghao Wu +14Vision-Language-Action ModelsRobotic Manipulation

  13. RAFT: Reasoning-Aligned Fine-Tuning Improves Robustness of VLA Post-Training

    Nov 3, 2026Shresth Grover, Andrey Kolobov, Dhruv ShahVLA AdaptationVision-Language-Action Models

  14. PREM: Prediction-Regulated Memory for Robot Manipulation

    Oct 31, 2026Xiao Huang, Ya Zhang, Weidi XieMemory-Augmented VLMsRobotic Manipulation

  15. Tokens of Motion: Primitive Motion Learning for Human-Humanoid Skill Transfer

    Oct 27, 2026Hanwen Ren, Meng Zheng, Wenzhe Cui +5VLA AdaptationRobustness of VLAs

  16. Retrieve to Act: Motion Primitive Graph Retrieval for Robust VLA Execution

    Oct 27, 2026Pham Tri Quang, Bao-Ngoc Dao, Minh Duc Nguyen +15Robustness of VLAsTest-Time Adaptation of VLAs

  17. EgoForesight: Effectively Scaling Policy Learning with Noisy Egocentric Actions

    Oct 22, 2026Chubin Zhang, Changyuan Wang, Yirui Wang +7Robot Imitation LearningRobot Learning from Video

  18. FIT-MAC: A Closed-Loop Multi-Agent Framework for Robust Vision-Language-Action Models

    Oct 20, 2026Jiatong Li, Jisheng Dang, Qiangqiang Ma +6Closed-Loop Control of VLAsRobustness of VLAs

  19. PACT-VLA: Speculative Action-Chunk Verification with Pre-Commit Observations

    Oct 20, 2026Yixiang Zhu, Ruilong Wu, Yonghao Chen +3Vision-Language-Action ModelsRobustness of VLAs

  20. RoboCasa-Shift: Stage-Aware Robustness Evaluation of Vision-Language-Action Policies in Long-Horizon Household Manipulation

    Oct 20, 2026Haocheng Wang, Youquan Liu, Weiyu Guo +6Robustness of VLAsRobotics

  21. VLA-Retain: Benchmarking Cross-Dimension Effects of Vision Tuning in VLAs

    Oct 20, 2026Borui Xu, Han Chen, Bingsheng He +3VLA AdaptationRobustness of VLAs

  22. What Would Removing This Corruption Change? Failure Attribution for Vision-Language-Action Models

    Oct 19, 2026Carson Sobolewski, Anjali Parashar, Fei Chen +2Robustness of VLAs

  23. LIBERO-Pro2.0: A Benchmark for Evaluating the Robustness of Embodied Foundation Models

    Oct 17, 2026Xueyang Zhou, Yangming Xu, Qianjiang Li +10Distribution Shift RobustnessEmbodied Agent Benchmarks

  24. Enhancing Visual Task-Redirection Attacks on Vision-Language-Action Models via Closed-Loop Optimization

    Oct 16, 2026Chunjin Ye, Guanying Chen, Jianjie Huang +6Adversarial Attacks on VLAsAdversarial Attacks

  25. CarryOn: A Dual-Loop Harness for Personalized and Reliable Robot Assistance

    Oct 15, 2026Zixuan Wang, Yuqi Liu, Yuxin Chen +7Assistive RoboticsRobot Safety