先學移動，再學執行：針對VLA的任務不可知預訓練

摘要

視覺-語言-行動（VLA）模型的根本瓶頸在於缺乏專家示範——即由觀測、指令與動作組成的三元組，此類數據在大規模收集時成本極高。我們認為，此瓶頸源於混淆了兩個不同的學習目標：獲取物理能力（如何移動）與獲取語義對齊（做什麼）。關鍵在於，僅後者需要語言監督。基於此分解假說，我們提出任務無關預訓練（TAP），這是一個兩階段框架：首先透過自監督逆向動力學目標，從廉價的未標記互動數據中學習可遷移的動作先驗——包括被捨棄的非任務軌跡與自主機器人遊玩數據。隨後，輕量級的第二階段使用極少量專家數據，將這些先驗與語言對齊。在SIMPLER基準上，TAP達到了與使用超過100萬條專家軌跡訓練的模型相當的表現，同時僅使用數量級更少的標記數據，比標準行為克隆實現了10%的絕對提升。在真實世界的WidowX平台上，TAP在相機擾動下仍保持25%的成功率，而網路規模基線則降至0%，證明任務無關預訓練能產生穩健且可遷移的物理表徵，並為具身人工智慧提供了可擴展的發展路徑。

English

Vision-Language-Action (VLA) models are fundamentally bottlenecked by the scarcity of expert demonstrations -- triplets of observations, instructions, and actions that are costly to collect at scale. We argue that this bottleneck stems from conflating two distinct learning objectives: acquiring physical competence (how to move) and acquiring semantic alignment (what to do). Crucially, only the latter requires language supervision. Building on this Decomposition Hypothesis, we propose Task-Agnostic Pretraining (TAP), a two-stage framework that first learns transferable motor priors from cheap, unlabeled interaction data -- including discarded off-task trajectories and autonomous robot play -- via a self-supervised Inverse Dynamics objective. A lightweight second stage then grounds these priors in language using minimal expert data. On the SIMPLER benchmark, TAP matches models trained on over 1M expert trajectories while using orders of magnitude less labeled data, yielding a 10% absolute gain over standard behavior cloning. On a real-world WidowX platform, TAP retains 25% success under camera perturbations where internet-scale baselines collapse to 0%, demonstrating that task-agnostic pretraining produces robust, transferable physical representations and offers a scalable path forward for Embodied AI.