外观
论文arXiv 2610.2236213 天前
CoDA：数据引擎与动作策略的协同进化
原文链接：CoDA: Co-Evolution of Data Engines and Action Policies（arXiv 2610.22362 · 发布日期：2026-10-15）
CoDA: Co-Evolution of Data Engines and Action Policies
CoDA让仿真数据引擎与动作策略通过共享记忆协同进化，并在操作任务与LIBERO-Pro上验证提升。
核心要点
- 动机是物理有效的生成环境未必能针对策略失败提供有用训练。
- 框架用异构记忆汇集环境构建、策略失败和学习进展，驱动生成与训练程序。
- Text2potential将诊断转为分阶段势函数，为动作块价值学习提供稠密监督。
- 四类操作任务中OOD成功率从5.3%升至27.6%；LIBERO-Pro扰动下也显著提升。
验证范围与局限：验证范围为四类操作任务和LIBERO-Pro的位置、指令扰动设置；摘要未说明真机实验。
英文摘要 (Abstract)
Generative simulation can automate robot environment construction, but physically valid environments are not necessarily useful for training. The challenge is to identify environments that address a policy's failures and supervision that helps it overcome them. We introduce CoDA, a framework in which a simulation data engine and an action policy co-evolve through a shared heterogeneous memory of environment construction, policy failures, and learning progress. The memory compiles validated diagnoses from a vision-language critic into generation programs that recreate difficulties and training programs that help the policy overcome them. We further propose Text2potential to convert these diagnoses into staged potentials for action-chunk value learning, providing dense supervision for policy updates while preserving the task's optimal policy under standard shaping assumptions. Measured learning progress guides further generation and program refinement. Across four manipulation task families, CoDA improves out-of-distribution success from 5.3% after supervised fine-tuning to 27.6%, exceeding the strongest closed-loop baseline by 9.5 percentage points with the same generation budget. On LIBERO-Pro, 15 rounds of generation and training raise success from 25% to 84% under position perturbations and from 1% to 76% under instruction perturbations.
作者：Tianjiao Yu, Zhilin Ou, Yifan Shen, Xinzhuo Li, Yuanzhe Liu, Onkar Susladkar, Fan Wang, Ismini Lourentzou
来源
本页事实对应的公开页面，以原文为准。