The paper investigates a specific failure mode in vision‑language‑action models, termed instruction‑action binding, where models respond to language and vision separately but fail to combine them to select the correct action under counterfactual changes. Through behavioral analyses of fine‑tuned policies, the authors show that failed rollouts often preserve source behavior or switch to other demonstrated tasks, indicating that language is not ignored but mis‑bound. They propose Equivariant Counterfactual Training (ECT), which supplies counterfactual demonstrations and a paired loss to enforce correct action selection, achieving significant performance gains across simulated and real‑world benchmarks.
By Hung-Jen Chen, Yu-Hsun Hou, Yan-Hong Chen, Yan-Fu Chen, Binghua Cai, Min Sun, Chun-Yi Lee
The paper investigates whether a single descriptive sentence can correct failures in frozen vision‑language‑action (VLA) policies. By labeling failures in the LIBERO‑90 benchmark, the authors identify four failure modes, finding that only execution misalignment—where the policy reaches the correct target but misjudges distance or force—can be remedied with a concise human‑generated sentence. This sentence, parsed by an LLM into a bias on the policy’s actions, successfully restores performance on both simulated and real‑world tasks, while automated language models fail to achieve similar gains.
By Owen Kwon, Pablo Ortega-Kral, Arthur Bucker, Jean Oh
arXiv:2608.23224v1 Announce Type: cross
Abstract: Retrieval can efficiently and effectively augment a frozen vision--language--action (VLA) policy without retraining, yet retrieved text becomes a con...
By Zhiruo Zhou, Zelin Li, Xiwen Chen, Jiazhuo Li, Chenwei Wang, Huiming Chen, Xiaojun Zhu
arXiv:2610.32941v1 Announce Type: cross
Abstract: Vision-Language-Action (VLA) models perform well on robotic manipulation, yet remain brittle once their own actions carry them into states their trai...
By Thomas Landeg, Thierry Tambe
arXiv:2610.34108v1 Announce Type: cross
Abstract: Vision-Language-Action (VLA) models map a language instruction and the current observation directly to actions, without checking whether the instruct...
By Jie Wang, Bo Zhang, Renzhi Lu, Jianyu Chen
arXiv:2610.00604v1 Announce Type: cross
Abstract: Vision-language-action policies often see only one or a few recent frames, which makes it difficult to evaluate how they use information that disappe...
By Egor Cherepanov, Nikita Kachaev, Aleksandr I. Panov, Alexey K. Kovalev
EmbodiedSkills is a unified framework that treats each skill decision as an execution proposal, checking prerequisites and verifying outcomes during long‑horizon vision‑language‑action tasks. It connects high‑level skill selection, bounded low‑level VLA execution, and post‑action verification through a fixed executable‑skill interface, enabling easy replacement of low‑level policies and recording of structured trajectories for supervision and adaptation. Instantiated with Qwen3‑VL and OpenPI/pi0.5 on RoboTwin 2.0 and LIBERO, the framework achieves high success rates (86.20% and 97.40% respectively) and demonstrates effective memory‑dependent task performance.
By Wei Wang, Wenqiao Zhang, Yutong Lin, Yuqian Yuan, Tianwei Lin, Jinhao Mao, Zhenxuan Fan, Mingjian Gao, Yang Dai, Wentong Li, Zheqi Lv, Zheng Dong, Yingjie Niu, Jiaqi Zhu, Jun Xiao, Chao Li, Yueting Zhuang
arXiv:2603.12717v2 Announce Type: replace-cross
Abstract: Vision-language-action policies map camera images and natural-language instructions to a robot's motor actions. Some of these policies are de...
By Tuan Duong Trinh, Basim Azam, Mohammed Ishaq Ansari, Mohammed Yaqoob Ansari, Naveed Akhtar
VLCP (Vision Language Control Policy) is a training‑free robot manipulation approach that keeps a vision‑language model (VLM) frozen and uses it to generate short Python control functions. Unlike traditional methods that retry a fixed policy, VLCP rewrites the control code every K steps based on multi‑view RGB, proprioceptive state, and state delta, allowing failures to be corrected within the same episode. In a 57‑task MuJoCo/RoboVerse benchmark, VLCP achieves 35.1% pooled success versus 3.5% for a single‑query baseline, with a 27.3% within‑episode recovery rate on failed grasps and efficient token usage.
By Dhia Naouali, Minghan Wu, Claudia Wong, Abhinav Puthran, Omar G. Younis
arXiv:2610.10526v1 Announce Type: cross
Abstract: Vision-language-action models (VLAs) are strikingly sensitive to instruction phrasing and do not inherit the language robustness of the vision-langua...
By Mikey Watts (Independent Researcher), Yuchen Cui (University of California, Los Angeles)
IndustrialVLA-Bench evaluates six released vision‑language‑action (VLA) and world‑action (WAM) robot policy models using a unified reporting schema. It measures clean capability on LIBERO, non‑language robustness on LIBERO‑Plus, instruction sensitivity on LIBERO‑Para, and observed execution cost, aggregating results across three random seeds. The benchmark also reports inference latency, peak memory, runtime mode, and evidence status for each system, enabling traceable comparisons without claiming one paradigm is universally superior.
By Yiqi Wang, Zhifeng Rao, Jiaqi Zhang, Xiaoyang Li, Zhangkai Wu, Yiqun Duan, Mingkai Zheng, Fei Wang, Shan You, Taotao Cai
arXiv:2610.05166v2 Announce Type: replace
Abstract: A safe action is not necessarily a viable one. A frozen vision-language-action (VLA) policy can favor a locally admissible move that leaves no poli...
By Tu Nguyen, Matthieu Zimmer, Vu Anh Vu, Ziyi Wang, Jannik Hammel Nielsen, Xuebing Zhou, Haitham Bou Ammar