让机器人部署后能像人一样从经验中学习，提升适应能力。
Dejavu: Towards Experience Feedback Learning for Embodied Intelligence
- 用记忆检索网络从过往经验中找相似场景的应对策略。
- 在多个任务中使成功率显著高于固定策略基线。
- 适合需要长期迭代优化的现实世界机器人应用。
具身智能体在真实环境中部署后难以获取新知识以提升性能。本文提出 Dejavu，一种通用的部署后学习框架，通过经验反馈网络（EFN）为冻结的视觉-语言-动作（VLA）策略注入检索到的执行记忆。EFN根据当前环境上下文，识别相关历史行为并以此指导动作预测。我们使用强化学习和语义相似性奖励训练 EFN，使预测动作与当前观测下过往行为保持一致。部署时，EFN持续扩展记忆库，实现“从经验中学习”。在多样化具身任务上的实验表明，该方法显著提升了适应性、鲁棒性和成功率。项目主页：https://dejavu2025.github.io/
原文摘要 · Abstract (English)
Embodied agents face a fundamental limitation: once deployed in real-world environments, they cannot easily acquire new knowledge to improve task performance. In this paper, we propose Dejavu, a general post-deployment learning framework that augments a frozen Vision-Language-Action (VLA) policy with retrieved execution memories through an Experience Feedback Network (EFN). EFN identifies contextually relevant prior action experiences and conditions action prediction on the retrieved guidance. We train EFN with reinforcement learning and semantic similarity rewards, encouraging the predicted actions to align with past behaviors under the current observation. During deployment, EFN continually expands its memory with new trajectories, enabling the agent to exhibit ``learning from experience.'' Experiments across diverse embodied tasks show that EFN improves adaptability, robustness, and success rates over frozen baselines. Our Project Page is https://dejavu2025.github.io/.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。