Search papers, labs, and topics across Lattice.
The paper introduces MemFly, a novel memory optimization framework for LLMs based on the information bottleneck principle, addressing the trade-off between memory compression and precise retrieval. MemFly uses a gradient-free optimizer to minimize compression entropy while maximizing relevance entropy, creating a stratified memory structure. Experiments demonstrate that MemFly outperforms existing methods in memory coherence, response fidelity, and accuracy by leveraging a hybrid retrieval mechanism combining semantic, symbolic, and topological pathways.
LLMs can now achieve better memory coherence and response fidelity thanks to MemFly's information bottleneck approach to on-the-fly memory optimization.
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.
