SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.03296 · ICML · 2026

PlugMem: A Task-Agnostic Plugin Memory Module for LLM Agents

Jianfeng Gao, Jiawei Han, Michel Galley, Chengxiang Zhai, Chenglong Wang, Zixi Chen, Xuan He, Ke Yang, Jize Jiang

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Long-term memory is essential for large language model (LLM) agents operating in complex environments, yet existing memory designs are either task-specific and non-transferable, or task-agnostic but less effective due to low task-relevance and context explosion from raw memory retrieval. We propose PLUGMEM, a task-agnostic plugin memory module that can be attached to arbitrary LLM agents without task-specific redesign. Motivated by the fact that decision-relevant information is concentrated as abstract knowledge rather than raw experience, we draw on cognitive science to structure episodic memories into a compact, extensible knowledge-centric memory graph that explicitly represents propositional and prescriptive knowledge. This representation enables efficient memory retrieval and reasoning over task-relevant knowledge, rather than verbose raw trajectories, and departs from other graph-based methods like GraphRAG by treating knowledge as the unit of memory access and organization instead of entities or text chunks. We evaluate PLUGMEM unchanged across three heterogeneous benchmarks (long-horizon conversational question answering, multi-hop knowledge retrieval, and web agent tasks). The results show that PLUGMEM consistently outperforms task-agnostic baselines and exceeds task-specific memory designs, while also achieving the highest information density under a unified information-theoretic analysis. Code and data are available at https://github.com/TIMANgroup/PlugMem. 1 Each point corresponds to a memory method evaluated on Num of Mem Token (Normalized) PMI (Normalized) PlugMem (LME) PlugMem (HPQA) PlugMem (WA) Vanilla Retrieval A-Mem LiCoMemory RAPTOR HippoRAG2 AWM Higher utility Lower cost Figure 1. A utility-cost visualization of agentic memory approaches. 1 PLUGMEM, evaluated unchanged across heterogeneous benchmarks requiring processing multiple memory types, achieves the highest decision-making utility of memory at the lowest agent-side memory cost.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2603.03296")
get_code_for_paper("2603.03296")
have("2603.03296")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2603.03296.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2603.03296.svg)](https://syntology.ai/paper/2603.03296)

Connect an agent — have() is free.