Code-as-Monitor: Constraint-aware Visual Programming for Reactive and Proactive Robotic Failure Detection
Enshen Zhou, Qi Su, Cheng Chi, Zhizheng Zhang, Zhongyuan Wang, Tiejun Huang, Lu Sheng, He Wang
摘要
Proactive Failure Detection (a) Reactive Failure Detection Task: Move the pan with the lobster to the stove, and be careful not to let the lobster drop out. (c) Real-world Test Pan Lobster Stove jump out Figure 1. For the task "Move the pan with lobster to the stove without losing the lobster", (a) reactive failure detection identifies failures after they occur, and (b) proactive failure detection prevents foreseeable failures. In (a), at t R 4 , the robot detects the failure after the lobster unpredictably jumps out due to the heat. In (b), pan tilting is detected at t P 3 and corrected it at t P ′ 3 , requiring real-time precision. We formulate both tasks as spatio-temporal constraint satisfaction problems, leveraging our proposed constraint elements for precise, real-time checking. For example, in (a), a large relative distance between pan and lobster indicates failure; in (b), a large angle between the pan and the horizontal plane needs correction. (c) shows that our method combined with an open-loop policy forms a closed-loop system, enabling proactive (e.g., detecting moving glass during grasping) and reactive (e.g., removing toy after grasping) failure detection in cluttered scenes.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper16
- RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for RoboticsEnshen Zhou, Jingkun An, Cheng Chi, Yi Han 等NeurIPS 2025 · 被引用 159 次
- SoFar: Language-Grounded Orientation Bridges Spatial Reasoning and Object ManipulationZekun Qi, Wenyao Zhang, Yufei Ding, Runpei Dong 等NeurIPS 2025 · 被引用 65 次
- Reason-RFT: Reinforcement Fine-Tuning for Visual Reasoning of Vision Language ModelsHuajie Tan, Yuheng Ji, Xiaoshuai Hao, Xiansheng Chen 等NeurIPS 2025 · 被引用 45 次
- From Language to Locomotion: Retargeting-free Humanoid Control via Motion Latent GuidanceZhe Li, Yangyang Wei, Boan Zhu, Yibo Peng 等ICLR 2026 · 被引用 29 次
- ManipLVM-R1: Reinforcement Learning for Reasoning in Embodied Manipulation with Large Vision-Language ModelsZirui Song, Guangxian Ouyang, Mingzhe Li, Yuheng Ji 等AAAI 2026 · 被引用 21 次
它引用的顶会 Paper19
- Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah 等NeurIPS 2020 · 被引用 64,255 次
- LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu 等ICLR 2022 · 被引用 18,833 次
- Visual Instruction TuningHaotian Liu, Chunyuan Li, Qingyang Wu, Yong Jae LeeNeurIPS 2023 · 被引用 11,349 次
- PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMsSoroush Nasiriany, Fei Xia, Wenhao Yu, Ted Xiao 等ICML 2024 · 被引用 212 次
- Generalized Planning in PDDL Domains with Pretrained Large Language ModelsTom Silver, Soham Dan, Kavitha Srinivas, Joshua B. Tenenbaum 等AAAI 2024 · 被引用 194 次
相关 Paper
- SAFE: Multitask Failure Detection for Vision-Language-Action ModelsQiao Gu, Yuanliang Ju, Shengxiang Sun, Igor Gilitschenski 等NeurIPS 2025 · 被引用 103 次
- RoboFailRing: Retrieval-Augmented and Language Grounding Failure Detection for VLM-enabled Robotic ManipulationChenduo Ying, Linkang Du, Yuanchao Shu, Peng ChengACL 2026
- Self-Correcting Robot Manipulation via Gaussian-Splatted ForesightShaohui Pan, Yong Xu, Ruotao Xu, Zihan Zhou 等AAAI 2025 · 被引用 2 次
- Self-Refining Vision Language Model for Robotic Failure Detection and ReasoningCarl Qi, Xiaojie Wang, Silong Yong, Stephen Sheng 等ICLR 2026 · 被引用 8 次
- PREGO: Online Mistake Detection in PRocedural EGOcentric VideosAlessandro Flaborea, Guido Maria D'Amely di Melendugno, Leonardo Plini, Luca Scofano 等CVPR 2024
