外观
论文arXiv 2610.0917022 天前
超越重建：机器人策略的动作标记化中什么才重要？
原文链接：Beyond Reconstruction: What Matters in Action Tokenization for Robot Policies?（arXiv 2610.09170 · 发布日期：2026-10-06）
Beyond Reconstruction: What Matters in Action Tokenization for Robot Policies?
针对动作标记化不只依赖重建的问题，ProAct联合提升可预测性与鲁棒性，并在多个基准、VLA和真机操作中验证。
核心要点
- 自回归动作token策略需将离散序列转为连续控制，单纯重建不足以保证成功。
- ProAct在重建外优化下游token可预测性和未见预测解码的鲁棒性。
- 该方法与策略无关，仅用动作数据训练，适用于多种tokenizer架构。
- 在多项基准、VLA和真机操作中，成功率平均提升11.3、21.8和36.7个百分点。
验证范围与局限：验证覆盖Robomimic、LIBERO、RoboTwin基准、VLA策略和真实世界机器人操作。
英文摘要 (Abstract)
Autoregressive action-token policies such as vision-language-action models require action tokenizers to translate discrete token sequences into precise control actions in continuous space. Many action tokenizers learn the mapping between tokens and actions via a reconstruction objective. However, as we show through extensive analysis, sufficiently accurate action reconstruction is only one part of what makes a downstream robot policy successful. It is also critical that the policy is able to predict the right tokens for new observations, and that unseen policy token predictions still decode into reasonable actions. These properties are downstream of tokenizer training and are not directly incentivized by a reconstruction objective alone. In this work, we introduce Predictable and Robust Action Tokenization (ProAct), a tokenizer training method that strategically augments reconstruction with the goal of improving downstream predictability and robustness. ProAct is policy-agnostic and uses only action datasets for training. Across the Robomimic, LIBERO, and RoboTwin benchmarks and a diverse set of tokenizer architectures, ProAct improves rollout success by an average of 11.3 percentage points. These improvements also translate to vision-language-action policies and real-world robotic manipulation, yielding average gains of 21.8 and 36.7 percentage points, respectively. These results suggest that effective action tokenization should be designed as a policy interface that balances fidelity, predictability, and robustness, rather than as a reconstruction problem alone.
作者：Haoran Chen, Jingtian Ji, Samuel Wheeler, Kaylene C. Stocking, Matthew R. Walter
来源
本页事实对应的公开页面，以原文为准。