# CAP: 用于令牌化机器人策略的码本对齐预测

> 原文链接：https://robotics.hellobike.com/papers/2026-10-06-2610-09178
> 频道：papers | 更新日期：2026-10-06 | 标签：arxiv, cs.RO
> 核心摘要：针对动作令牌化策略忽视码本结构的问题，CAP复用量化器码向量作类别原型，并在仿真基准和真机任务验证提升成功率。

# CAP: 用于令牌化机器人策略的码本对齐预测

::: v-pre
> 原文链接：[CAP: Codebook-Aligned Prediction for Tokenized Robot Policies](https://arxiv.org/abs/2610.09178)（arXiv 2610.09178 · 发布日期：2026-10-06）

**CAP: Codebook-Aligned Prediction for Tokenized Robot Policies**

针对动作令牌化策略忽视码本结构的问题，CAP复用量化器码向量作类别原型，并在仿真基准和真机任务验证提升成功率。

## 核心要点

- 动作令牌化将连续动作离散化，但现有策略常把令牌当作无关类别索引。
- CAP直接复用分词器码向量作为策略类别原型，其余分词器和策略骨干保持不变。
- 在四类量化器、三个仿真基准和两个真机任务上，CAP相较标准分类头持续提升任务成功率。
- 分析表明收益并非来自更高令牌准确率，而是码本结构让动作空间错误更温和并改善骨干表征。

**验证范围与局限**：验证范围为三个仿真基准和两个真实机器人任务，覆盖四类量化器。

## 英文摘要 (Abstract)

Action tokenization converts continuous robot actions into discrete symbols that can be modeled autoregressively. However, existing tokenizer-based policies typically ignore the tokenizer's learned latent code structure: after tokenization, the policy treats tokens as unrelated class indices and learns a new classifier from scratch. We show that this discarded structure is valuable. We introduce Codebook-Aligned Prediction (CAP), a method that directly reuses the tokenizer's code vectors as policy class prototypes while leaving the tokenizer and policy backbone otherwise unchanged. Across four quantizer families, three simulation benchmarks, and two real-robot tasks, CAP consistently improves task success over standard token classification heads while holding the tokenizer (and therefore its reconstruction quality) fixed. Our analysis further shows that these gains are not explained by higher token accuracy or changes in the policy head alone. Instead, reusing the tokenizer codebook provides the policy with valuable information about the tokenizer's learned latent structure across tokens, making token prediction errors more benign in action space and improving the representations learned by the policy backbone. These results suggest that action tokenizers learn useful action-aware latent structure beyond discrete targets that should be preserved when training downstream policies.

作者：Haoran Chen, Jingtian Ji, Samuel Wheeler, Kaylene C. Stocking, Matthew R. Walter
:::

## 原始参考来源 (Sources)

- [arXiv 2610.09178：CAP: Codebook-Aligned Prediction for Tokenized Robot Policies](https://arxiv.org/abs/2610.09178)