外观
论文arXiv 2610.2337812 天前
行动前先观察：通过观察对齐数据适配连接主动感知与视觉-语言-动作模型
原文链接：Look Before You Act: Bridging Active Perception and Vision-Language-Action Models via Observation-Aligned Data Adaptation（arXiv 2610.23378 · 发布日期：2026-10-16）
Look Before You Act: Bridging Active Perception and Vision-Language-Action Models via Observation-Aligned Data Adaptation
提出ActiveAlign，使VLA在主动感知终止状态上训练，并在LIBERO任务和真实遮挡场景验证成功率提升。
核心要点
- 杂乱操控需决定看哪里；直接加相机重定位会引发首帧分布敏感并降低成功率。
- ActiveAlign用视点模块生成相机轨迹，离线闭环对齐到演示并保留成功续段微调VLA。
- 在40个LIBERO任务上，pi_0和pi_0.5分别达到70.7%和72.7%成功率，优于标准训练基线。
- 匹配状态和数据匹配对照支持收益；OpenVLA与GR00T N1也显示相同趋势。
验证范围与局限：验证范围包括40个LIBERO任务和真实世界物理遮挡下的90次试验。
英文摘要 (Abstract)
Robots manipulating objects in clutter must decide not only how to act, but also where to look. Although active camera motion can expose an occluded target, it may move the robot into an initial observation-state distribution that differs from the manipulation policy's training distribution. We identify this problem as first-frame distribution sensitivity: under perturbed starts in LIBERO, naively prepending camera repositioning to pi_0.5 reduces success from 46.3% to 11.5%. We introduce ActiveAlign, a data-adaptation framework that trains a Vision-Language-Action (VLA) policy from the states at which active perception terminates. A learned viewpoint module first generates a camera trajectory. Offline, ActiveAlign uses closed-loop alignment to connect the trajectory endpoint to an existing demonstration, replays the remaining manipulation trajectory, and retains successful continuations for fine-tuning. At deployment, the same viewpoint module precedes the adapted VLA without modifying its architecture. Across 40 LIBERO tasks, ActiveAlign achieves 70.7% success with pi_0 and 72.7% with pi_0.5, outperforming their standard-training baselines by 40.0% and 26.4% respectively. Matched-state and data-matched controls support gains beyond prefix shortening and generic viewpoint augmentation; OpenVLA and GR00T N1 also show the same trend. Success rate under physical occlusion in real world increases from 35/90 to 83/90. These results show that active perception benefits manipulation when policy training is aligned with the observation and robot-state distribution induced by viewpoint selection.
作者：Jiawei Geng, Zhenyang Liu, Yuzheng Wu, Yongchong Gu, Yanwei Fu
来源
本页事实对应的公开页面，以原文为准。