Long-term memory enables large language model agents to tackle complex tasks through historical interactions. However, existing frameworks encounter a fundamental dilemma between compressing redundant information efficiently and maintaining precise retrieval for downstream tasks. To bridge this gap, we propose MemFly, a framework grounded in information bottleneck principles that facilitates on-the-fly memory evolution for LLMs. Our approach minimizes compression entropy while maximizing relevance entropy via a gradient-free optimizer, constructing a stratified memory structure for efficient storage. To fully leverage MemFly, we develop a hybrid retrieval mechanism that seamlessly integrates semantic, symbolic, and topological pathways, incorporating iterative refinement to handle complex multi-hop queries. Comprehensive experiments demonstrate that MemFly substantially outperforms state-of-the-art baselines in memory coherence, response fidelity, and accuracy.
@article{arxiv.2602.07885,
title = {MemFly: On-the-Fly Memory Optimization via Information Bottleneck},
author = {Zhenyuan Zhang and Xianzhang Jia and Zhiqin Yang and Zhenbo Song and Wei Xue and Sirui Han and Yike Guo},
journal= {arXiv preprint arXiv:2602.07885},
year = {2026}
}