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. H-VLA: Hierarchical Vision-Language-Action Model with Key-Action Reasoning and Motion Planning in a Unified Action Space

    Sep 19, 2026Xiongfeng Peng, Lu Xu, Yandong Wang +10Vision-Language-Action ModelsLanguage-Conditioned Robot Manipulation

  2. VLA-Scope: Shift-Aware Failure Prediction for Vision-Language-Action Models

    Sep 18, 2026Kaiwen Zhu, Dongfang Liu, Liangkai LiuFailure Prediction for VLAsFailure Detection for VLAs

  3. Beyond Patch Removal: Persistent Adversarial Effects in Vision-Language-Action Policies

    Sep 17, 2026Enhao Wu, Fusen Guo, Yuxin Cao +3Adversarial Patch AttacksAdversarial Attacks on VLAs

  4. SAVLA: Symmetry-Aware Vision-Language-Action Models for Robotic Manipulation

    Sep 15, 2026Junle Li, Weixian Waylon Li, Fuxiang Wu +2Language-Conditioned Robot ManipulationVision-Language-Action Models

  5. Validating, Not Sampling: Region-Level Robustness of Vision-Language and Vision-Language-Action Models

    Sep 14, 2026Bogdan Aron, Christopher Brix, Benedikt Brückner +3VLM RobustnessRobustness of VLAs

  6. Atomic Motion Coordinate for Language-Steerable and Force-Responsive Manipulation

    Sep 14, 2026Jiaqi Zhai, Jingkai Zhao, Chen Yang +10Language-Conditioned Robot ManipulationForce-Controlled Robotic Manipulation

  7. Breaking the Vision-Action Shortcut: Latent Interface Training for Generalizable Robotics Foundation Models

    Sep 11, 2026Jianman Lin, Shailesh Shailesh, Zhongyi Luo +1Robot Foundation ModelsVision-Language-Action Models

  8. HuRo: Robotizing Human Videos for Scalable VLA Pretraining

    Sep 9, 2026Jinho Jeong, Se June Joo, Jaehyun Kang +4Cross-Embodiment Robot LearningVLA Training

  9. GloVLA: Let Geometry Move and Local VLA Interact for Robust Object-Centric Manipulation in Unstructured Environments

    Sep 5, 2026Truong Thanh Nguyen, Huy Hoang Nguyen, Ha Anh Nguyen +5Hierarchical Robot ControlRobotic Manipulation

  10. Where Success Breaks: Failure-Boundary Learning for Robust Vision-Language-Action Models

    Sep 5, 2026Yanzhe Chen, Zhijun Cao, Mike Zheng ShouVision-Language-Action ModelsRobustness of VLAs

  11. CR-VLA-Force: Learning Control-aware Compliance VLA Model for Robust Contact-rich Robotic Manipulation

    Sep 5, 2026Zhaohong Mai, Chao Wang, Chao Zeng +4Contact-Rich Robotic ManipulationForce-Controlled Robotic Manipulation

  12. LIBERO-RECOVER: Beyond Task Success Towards Failure Recovery in Robotic Manipulation Models

    Sep 4, 2026Lin Liu, Zhicheng Bao, Lu Zhang +7Robot Failure RecoveryRobotic Manipulation

  13. WISE: World-model-guided Imagination Scheduling for Efficient Post-training of Vision-Language-Action Models

    Sep 3, 2026Chenhao Zhang, Hanyu Zhao, Zihan Que +3Robustness of VLAsWorld Models for Robotics

  14. Sensing Which Modality Matters: Evidence-Gated Regularization for Robust VLA Policies

    Sep 2, 2026Yue Yang, Diego Romeres, Chiori Hori +3Robustness of VLAsMultimodal Sensor Fusion

  15. What Can One Sentence Fix? A Failure Taxonomy and a Minimal Language Correction for Frozen VLA Policies

    Aug 30, 2026Owen Kwon, Pablo Ortega-Kral, Arthur Bucker +1Vision-Language-Action ModelsTest-Time Adaptation of VLAs

  16. FLARE: A Failure-Aware Framework for Autonomous Correction and Recovery in Visual-Language Robotic Manipulation

    Aug 27, 2026Ganlong Zhao, Zijia Tang, Xingping Chen +3Robot Failure RecoveryRobustness of VLAs

  17. StreamPI: Streaming Multimodal Temporal Modeling for Vision-Language-Action Models

    Aug 26, 2026Zhe Liu, Jinghua Hou, Yuxiang Lu +7Temporal Modeling for VLAsVision-Language-Action Models

  18. Think Only When Needed: Prompt-Authority Control for Selective Slow-Path Intervention in Vision-Language-Action Manipulation

    Aug 24, 2026Zhiruo Zhou, Zelin Li, Xiwen Chen +4Vision-Language-Action ModelsVLMs for Robotics

  19. Robust Bimanual Vision-Language-Action Models via Embarrassingly Simple Modality Masking

    Aug 23, 2026Dongzhou Cheng, Ziang Li, Yixiao Zhou +6Vision-Language-Action ModelsRobustness of VLAs

  20. CounterAlign: Counterfactual Supervision for Vision-Language-Action Models

    Aug 22, 2026Haru Kondoh, Kei Ota, Asako Kanezaki +1RL for VLMsRobustness of VLAs

  21. GS-VLA: Plug-and-Play Viewpoint Canonicalization for Frozen VLA Policies via Gaussian Splatting

    Aug 19, 2026Yechan Park, HyunJin KimRobustness of VLAs

  22. Robo-Dopamine 2.0: History-Conditioned and OOD-Aware Process Reward Modeling for Robotic Manipulation

    Aug 16, 2026Yijie Xu, Haopeng Jin, Run Zhou +8Process Reward ModelsMultimodal Reward Modeling

  23. Bit-Flip Attacks on Vision-Language-Action Models: Action-Decoding Architecture Shapes the Vulnerability

    Aug 16, 2026Yudong Gao, Linghan Chen, Wenhan Wu +5Adversarial AttacksAdversarial Attacks on VLAs