# SGMem: Sentence Graph Memory for Long-Term Conversational Agents

Source: [SGMem: Sentence Graph Memory for Long-Term Conversational Agents](<https://www.sota2.com/research/paper/68fa7b8f1e209131cea891f2>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T05:43:27.824Z

Authors: Yaxiong Wu, Yongyue Zhang, Sheng Liang, Yong Liu

Published: 2025-09-25T14:21:44.000Z

## Abstract

Long-term conversational agents require effective memory management to handle dialogue histories that exceed the context window of large language models (LLMs). Existing methods based on fact extraction or summarization reduce redundancy but struggle to organize and retrieve relevant information across different granularities of dialogue and generated memory. We introduce SGMem (Sentence Graph Memory), which represents dialogue as sentence-level graphs within chunked units, capturing associations across turn-, round-, and session-level contexts. By combining retrieved raw dialogue with generated memory such as summaries, facts and insights, SGMem supplies LLMs with coherent and relevant context for response generation. Experiments on LongMemEval and LoCoMo show that SGMem consistently improves accuracy and outperforms strong baselines in long-term conversational question answering.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2509.21212>)

[PDF](<https://arxiv.org/pdf/2509.21212.pdf>)
