cs.RONov 8, 2026

Review the Episodes, Refine the Lessons: Simple Memory Evolution for Embodied Agents

Authors: Yifei Yao, Yude Zou, Xing Gao, Guanbo Jiao, Yukai Wang, Jianming Ma, Weinan Zhang, Chunhua Shen

Organizations: Shanghai Artificial Intelligence Laboratory · Shanghai Jiao Tong University · Fudan University · Zhejiang University

Abstract

Recent studies of embodied agents based on frontier multimodal large language models (MLLMs) develop harness engineering around the MLLM, such as harnesses with post-trained policies and recursive self-improvement loops. Yet whether these additions exceed the native capability of the MLLM remains unexplored. We introduce SimplEvo to explore how far an MLLM can improve under direct control when it retains its execution experience in self-written text memories. It requires no weight updates, post-trained policies, or privileged information. The MLLM writes a coordinate-free task memory once from a single demonstration. Under a strict train-test protocol, it then creates a bounded working memory on training episodes only and refines it after each episode with lessons on procedure, completion, and recovery. Both memories stay fixed at test time. On EBench, SimplEvo lifts the success rate (SR) of GPT-6-Astra under direct control from 46.7% to 81.7%, and tops the RoboDojo leaderboard with 69.63% SR and a score of 75.32. Compared with memories that also store past visual frames, SimplEvo reaches a similar SR with text memory alone while using fewer tokens. With its memories fixed, SimplEvo also generalizes to unseen layouts and randomized scene configurations. Moreover, the working memory transfers across models: written by GPT-6-Astra, it can be reused as is by weaker MLLMs and improve their performance without any adaptation of their own.

Figures & tables

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. What Should Embodied Agents Remember? Benchmarking and Evaluating Memory Modality for Embodied Agents

    Oct 8, 2026Junxiang Wang, Yuchen Liu, Yi Wu +6Long-Term Memory for Embodied AgentsAgent Memory

  2. HIMM: Human-Inspired Long-Term Memory Modeling for Embodied Exploration and Question Answering

    Feb 17, 2026Ji Li, Bo Wang, Jing Xia +2Long-Term Memory for Embodied AgentsMemory-Augmented Agents

  3. MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents

    Feb 2, 2026Haozhen Zhang, Quanyu Long, Jianzhu Bao +4Memory Management for LLM AgentsMemory Consolidation for LLM Agents