Bi-Mem: Bidirectional Construction of Hierarchical Memory for Personalized LLMs via Inductive-Reflective Agents

2026-01-10

arXiv.org

AI Summary

This work addresses the distortion of user profiles in long-term personalized dialogue caused by context window limitations and memory noise in large language models. To mitigate this, the authors propose an inductive-reflection dual-agent framework that constructs a hierarchical memory structure with bidirectional alignment across factual, situational, and personality layers. The inductive agent extracts factual information and infers user profiles, while the reflective agent calibrates local situational memories using global personality constraints to ensure consistency between global and local representations. Additionally, the framework incorporates a graph clustering–based organization of situational memories and an associative recall mechanism driven by diffusion-based activation. Experimental results demonstrate that the proposed approach significantly improves response accuracy in long-term personalized dialogue tasks, confirming its effectiveness and robustness in high-fidelity memory modeling and retrieval.

Technology Category

• Multiagent Systems: Adversarial Agents
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Natural Language Processing: (Large) Language Models

Application Category

• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Constructing memory from users'long-term conversations overcomes LLMs'contextual limitations and enables personalized interactions. Recent studies focus on hierarchical memory to model users'multi-granular behavioral patterns via clustering and aggregating historical conversations. However, conversational noise and memory hallucinations can be amplified during clustering, causing locally aggregated memories to misalign with the user's global persona. To mitigate this issue, we propose Bi-Mem, an agentic framework ensuring hierarchical memory fidelity through bidirectional construction. Specifically, we deploy an inductive agent to form the hierarchical memory: it extracts factual information from raw conversations to form fact-level memory, aggregates them into thematic scenes (i.e., local scene-level memory) using graph clustering, and infers users'profiles as global persona-level memory. Simultaneously, a reflective agent is designed to calibrate local scene-level memories using global constraints derived from the persona-level memory, thereby enforcing global-local alignment. For coherent memory recall, we propose an associative retrieval mechanism: beyond initial hierarchical search, a spreading activation process allows facts to evoke contextual scenes, while scene-level matches retrieve salient supporting factual information. Empirical evaluations demonstrate that Bi-Mem achieves significant improvements in question answering performance on long-term personalized conversational tasks.

Problem

• hierarchical memory
• memory hallucination
• personalized LLMs
• conversational noise
• global-local alignment

Innovation

• bidirectional memory construction
• inductive-reflective agents
• hierarchical memory
• global-local alignment
• associative retrieval