SimpleMem: Efficient Lifelong Memory for LLM Agents

Papers with Code Trending Papers
lifelong-memory llm-agents semantic-compression memory-framework efficient-retrieval token-efficiency

Summary

Introduces SimpleMem, an efficient memory framework for LLM agents that uses semantic lossless compression to improve accuracy and reduce token consumption, achieving 26.4% F1 improvement and up to 30x reduction in inference-time token usage.

To support reliable long-term interaction in complex environments, LLM agents require memory systems that efficiently manage historical experiences. Existing approaches either retain full interaction histories via passive context extension, leading to substantial redundancy, or rely on iterative reasoning to filter noise, incurring high token costs. To address this challenge, we introduce SimpleMem, an efficient memory framework based on semantic lossless compression. We propose a three-stage pipeline designed to maximize information density and token utilization: (1) Semantic Structured Compression, which applies entropy-aware filtering to distill unstructured interactions into compact, multi-view indexed memory units; (2) Recursive Memory Consolidation, an asynchronous process that integrates related units into higher-level abstract representations to reduce redundancy; and (3) Adaptive Query-Aware Retrieval, which dynamically adjusts retrieval scope based on query complexity to construct precise context efficiently. Experiments on benchmark datasets show that our method consistently outperforms baseline approaches in accuracy, retrieval efficiency, and inference cost, achieving an average F1 improvement of 26.4% while reducing inference-time token consumption by up to 30-fold, demonstrating a superior balance between performance and efficiency. Code is available at https://github.com/aiming-lab/SimpleMem.
Original Article
View Cached Full Text

Cached at: 05/24/26, 12:26 AM

Paper page - SimpleMem: Efficient Lifelong Memory for LLM Agents

Source: https://huggingface.co/papers/2601.02553

Abstract

Tosupportreliablelong-terminteractionincomplexenvironments,LLMagentsrequirememorysystemsthatefficientlymanagehistoricalexperiences.Existingapproacheseitherretainfullinteractionhistoriesviapassivecontextextension,leadingtosubstantialredundancy,orrelyoniterativereasoningtofilternoise,incurringhightokencosts.Toaddressthischallenge,weintroduceSimpleMem,anefficientmemoryframeworkbasedonsemanticlosslesscompression.Weproposeathree-stagepipelinedesignedtomaximizeinformationdensityandtokenutilization:(1)SemanticStructuredCompression,whichappliesentropy-awarefilteringtodistillunstructuredinteractionsintocompact,multi-viewindexedmemoryunits;(2)RecursiveMemoryConsolidation,anasynchronousprocessthatintegratesrelatedunitsintohigher-levelabstractrepresentationstoreduceredundancy;and(3)AdaptiveQuery-AwareRetrieval,whichdynamicallyadjustsretrievalscopebasedonquerycomplexitytoconstructprecisecontextefficiently.Experimentsonbenchmarkdatasetsshowthatourmethodconsistentlyoutperformsbaselineapproachesinaccuracy,retrievalefficiency,andinferencecost,achievinganaverageF1improvementof26.4%whilereducinginference-timetokenconsumptionbyupto30-fold,demonstratingasuperiorbalancebetweenperformanceandefficiency.Codeisavailableathttps://github.com/aiming-lab/SimpleMem.

View arXiv pageView PDFProject pageGitHub3.39kAdd to collection

Get this paper in your agent:

hf papers read 2601\.02553

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2601.02553 in a model README.md to link it from this page.

Datasets citing this paper2

#### molmohsen/awesome-ai-agent-papers #### zhongweixie/A-Survey-on-AI-Agent-Harness Viewer• Updated2 days ago • 1 • 28

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2601.02553 in a Space README.md to link it from this page.

Collections including this paper4

Similar Articles

MemRefine: LLM-Guided Compression for Long-Term Agent Memory

arXiv cs.CL

MemRefine is an LLM-guided framework for compressing long-term agent memory under fixed storage budgets, using similarity for candidate pairing and an LLM judge for factual deletion/merge decisions, outperforming rule-based baselines on benchmarks.