AI 看懂了物体在哪，却还是不会动手
给 AI 看一张图、问它东西在哪，它几乎满分；让它真的伸手去拿，成功率直接腰斩。这篇新基准 VA-Bench 把「看」和「做」拆开测：模型只能靠看视频学步骤、自己选摄像头角度、自己发坐标指令，没有作弊数据。结果最强的模型，定位物体 100% 对，但完整完成任务只有 54%;让它自己转视角比固定视角强一倍；换个没见过的房间布局，成功率掉 30 个点以上；五步的连续任务，没有一个模型能走完。它不是你明天能用上的东西，但它划了条线：现在的 AI 是很好的「眼睛」，却是很差的「手」。
📄 原文摘要(英文)
Spatial intelligence requires more than describing object locations. Under incomplete observation, models must identify and acquire missing evidence, interpret it in a common spatial frame, and act on it. We introduce VA-Bench to evaluate the complete observe-reason-act-revise loop. General-purpose MLLMs learn procedural context from RGB-only demonstrations, actively select camera viewpoints, issue metric Cartesian commands, and revise them from execution feedback. Models receive no privileged object poses, oracle trajectories, or learned action heads. A fixed model-agnostic controller executes only model-specified targets. VA-Bench contains 14 base task families (11 single-arm and three dual-arm), seven held-out geometry/layout variants, and a long-horizon five-object composition track. We evaluate 12 primary model conditions in three independent runs over the same 20 physically verified seeds per base task, reporting terminal success, nine trajectory-level behavioral diagnostics, and subtask progress. First, the best-performing model scores 100.0% on target localization and 78.9% on spatial relations in the annotated run. Its three-run macro-average task success is only 53.93+/-3.17%. Second, active camera control significantly improves task success over passive multi-view observation. In one matched comparison, success rises from 27.86% to 57.50%. Third, held-out geometric transfer can reduce task success by over 30 percentage points. No model completes a strict long-horizon episode, despite substantial partial progress. VA-Bench thus tests whether general-purpose MLLMs can turn visual demonstrations and actively acquired evidence into successful embodied action.