pith. sign in

arxiv: 2605.28773 · v1 · pith:ZGSJA7BHnew · submitted 2026-05-27 · 💻 cs.CL · cs.AI· cs.LG· cs.MA· cs.MM

Rethinking Memory as Continuously Evolving Connectivity

classification 💻 cs.CL cs.AIcs.LGcs.MAcs.MM
keywords memoryfluxmemagenticcontinuouslyenvironmentsheterogeneousshouldthree
0
0 comments X
read the original abstract

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.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.