‹ back to 2026-07-30

MemHarness: Memory Is Reconstructed, Not Replayed

Research LLM Agents

Ranking

Overall 86

Content 95

Popularity 65

Observed public metrics from 1 member.

Representative image for MemHarness: Memory Is Reconstructed, Not Replayed

Merged summary

TL;DR - MemHarness trains LLM agents to reconstruct retrieved experiences for their current state rather than replaying static memories verbatim. This reduces negative transfer and improves performance and out-of-distribution robustness.

A unified policy critiques and adapts retrieved experiences into context-grounded guidance at each decision step.
The reconstructive behavior emerges through end-to-end GRPO training.
On ALFWorld and WebShop, MemHarness outperforms pure reinforcement learning and static memory-augmented baselines.
Analyses suggest reconstruction also provides latent training guidance that improves intrinsic reasoning.

Sources (1)

MemHarness: Memory Is Reconstructed, Not Replayed

arXiv cs.AI Rong Wu, Daocheng Fu, Licheng Wen, Xuemeng Yang, Shu Zou, Jianbiao Mei, Yuxin Wang, Hairong Zhang, Yu Yang, Tao Hu, Cong Zhang, Botian Shi, Pinlong Cai 2026-07-30 arXiv:2607.28272

Public signals Hugging Face upvotes 17

Providers: Hugging Face · Upvotes 17 OpenAlex · N/A Publisher · N/A Semantic Scholar · N/A X · N/A Fetched 2026-08-28 14:31:40.900155 UTC

A unified policy critiques and adapts retrieved experiences into context-grounded guidance at each decision step.
The reconstructive behavior emerges through end-to-end GRPO training.
On ALFWorld and WebShop, MemHarness outperforms pure reinforcement learning and static memory-augmented baselines.
Analyses suggest reconstruction also provides latent training guidance that improves intrinsic reasoning.

item →