Existing memory-augmented LLM agents often treat memory as a static repository with pre-defined representations and fixed retrieval pipelines, which is brittle in dynamic agentic environments where feedback, task variation, and heterogeneous signals continuously reshape what should be remembered and how it should be connected. To address this, we propose FluxMem, a connectivity-evolving memory framework that models memory as a heterogeneous graph and progressively refines its topology through three stages: initial connection formation, feedback-driven refinement, and long-term consolidation. During execution, FluxMem repairs missing links, prunes interference, aligns abstraction granularity, and distills recurrent successful trajectories into reusable procedural circuits, guided by one metric for memory generalizability and evolutionary maturity. Across three fundamentally distinct benchmarks including LoCoMo, Mind2Web, and GAIA, FluxMem achieves consistent state-of-the-art performance, demonstrating strong adaptation and generalization in complex agentic environments. The code will be open-sourced in https://github.com/zjunlp/LightMem.
@article{arxiv.2605.28773,
title = {Rethinking Memory as Continuously Evolving Connectivity},
author = {Jizhan Fang and Buqiang Xu and Zhixian Wang and Haoliang Cao and Xinle Deng and Baohua Dong and Hangcheng Zhu and Ruohui Huang and Gang Yu and Ying Wei and Guozhou Zheng and Feiyu Xiong and Haofen Wang and Huajun Chen and Ningyu Zhang},
journal= {arXiv preprint arXiv:2605.28773},
year = {2026}
}