# Does Memory Need Graphs? A Unified Framework and Empirical Analysis for Long-Term Dialog Memory

Source: [Does Memory Need Graphs? A Unified Framework and Empirical Analysis for Long-Term Dialog Memory](<https://www.sota2.com/research/paper/695c96e053fdd5c3617433a3>)

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

Data updated: 2026-02-12T05:00:14.659Z

Authors: Sen Hu, Yuxiang Wei, Jiaxin Ran, Zhiyuan Yao, Xueran Han, Huacan Wang, Ronghao Chen, Lei Zou

Published: 2026-01-03T20:39:39.000Z

## Abstract

Graph structures are increasingly used in dialog memory systems, but empirical findings on their effectiveness remain inconsistent, making it unclear which design choices truly matter. We present an experimental, system-oriented analysis of long-term dialog memory architectures. We introduce a unified framework that decomposes dialog memory systems into core components and supports both graph-based and non-graph approaches. Under this framework, we conduct controlled, stage-wise experiments on LongMemEval and HaluMem, comparing common design choices in memory representation, organization, maintenance, and retrieval. Our results show that many performance differences are driven by foundational system settings rather than specific architectural innovations. Based on these findings, we identify stable and reliable strong baselines for future dialog memory research. Code are available at https://github.com/AvatarMemory/UnifiedMem

## 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.

- [Embodied Planning on WorldLines 21 action-dense samples downstream planning probe](<https://www.sota2.com/research/sota/embodied-planning-on-worldlines-21-action-dense-samples-downstream-planning-probe.md>): Embodied Planning; WorldLines 21 action-dense samples downstream planning probe; Plan Judge: 48.1.

- [Embodied Task Planning on WorldLines Downstream Embodied (planning set)](<https://www.sota2.com/research/sota/embodied-task-planning-on-worldlines-downstream-embodied-planning-set.md>): Embodied Task Planning; WorldLines Downstream Embodied (planning set); Plan Judge Score: 48.1.

- [Memory Question Answering on WorldLines Memory QA samples](<https://www.sota2.com/research/sota/memory-question-answering-on-worldlines-memory-qa-samples.md>): Memory Question Answering; WorldLines Memory QA samples; Judge Score: 45.7.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2601.01280>)

[PDF](<https://arxiv.org/pdf/2601.01280.pdf>)