AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

TransMem: Transition-Aware Retrieval for Evolving Personal Memory

Shigeng Chen, Linhao Luo, Changlong Shi, Zhangchi Qiu, Shirui Pan

NeurIPS 2026 · Large language models › Retrieval and knowledge · approach: Retrieval-augmented reasoning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes TransMem, a transition-aware retrieval approach for evolving personal memory that links how earlier preferences are refined, contradicted, or superseded by later evidence.

Problem

Long-term personalized agents must answer queries over user histories where preferences, goals, and constraints evolve across interleaved conversations; existing retrieval and structured-memory systems return isolated memory units without explaining how earlier preferences changed.

Approach

Retrieval that is aware of transitions between memory states (refinement, contradiction, supersession) so that returned memory explains which earlier preference was updated by later evidence, rather than isolated memory units.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add transition records and adaptive expansion over factual anchors only (ablation)TransMem · PersonaMem (4 settings)overall accuracyImproved
add intra-topic transition edges to factual anchors (ablation)TransMem · PersonaMem-32k Qwen3-8Boverall accuracy (pp): +0.42 ppImproved
add inter-topic transition edges to factual anchors (ablation)TransMem · PersonaMem-32k Qwen3-8Boverall accuracy (pp): +1.05 ppImproved
use both intra and inter-topic edges instead of one type (ablation)TransMem · PersonaMem (4 settings)overall accuracyImproved
use graph-grounded transition expansion instead of chronological neighbors (ablation)TransMem · PersonaMem-32k / 128k Qwen3-8Boverall accuracy: 53.47 -> 56.84 / 46.51 -> 48.03Improved
use graph expansion instead of flat transition retrieval (ablation)TransMem · PersonaMem-32k / 128k Qwen3-8Boverall accuracy: 54.32 -> 56.84 / 46.92 -> 48.03Improved
use adaptive stopping instead of best fixed hop depth (ablation)TransMem · PersonaMem-32k / 128k Qwen3-8Boverall accuracy: +1.68 pp / +0.83 ppImproved
add transition context for New Ideas questions (ablation)TransMem · PersonaMem-128k Qwen3-8BNew Ideas accuracy: -3.27 ppHurt
add transition records for Aligned Recommendations (ablation)TransMem · PersonaMem-128k Qwen3-8BAligned Recs accuracy: unchangedNo clear effect
compare retrieval token cost with A-MemTransMem · PersonaMem-32k Qwen3-8Binference tokens / latency: 3.4k -> 1.82k tokens; 2.35 s -> 0.98 sImproved

Experimental setup

Baselines: retrieval-based memory systems structured-memory systems

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy