Disentangling Long-Term Memory via Latent Neuro-Symbolic Reasoning
cs.AI, cs.CL, cs.LG
Submitted: 2026-09-16
Updated: 2026-09-18
License: http://creativecommons.org/licenses/by/4.0/
The gist: Personalized agents are required to reason over long-term history interactions to infer both explicit preferences and implicit behavioral evidence.
Terminology
Abstract
Personalized agents are required to reason over long-term history interactions to infer both explicit preferences and implicit behavioral evidence. While early flat retrieval methods score memory fragments independently and neglect the distributed information, current structured memory frameworks rely on query-agnostic static graphs that fail to capture the context-dependent relations. Crucially, raw textual memories are inherently entangled and noisy, making fine-grained personalization and cross-session reasoning computationally prohibitive. To this end, we present LGM, a novel neuro-symbolic framework that shifts long-term memory disentanglement into a continuous latent space. Specifically, (i) instead of persisting fixed graphs, we design a tailored latent graph construction with a sparse autoencoder. Subject to each query, it maps historical interactions into latent memory nodes and disentangles the memory traces into sparse concept activations, dynamically synthesizing query-aware relational edge weights. (ii) A graph encoder then treats the query embedding as a conditioning preference to direct non-linear message passing across the task-specific latent subgraph. This yields a highly expressive memory representation for effective activations. Extensive experiments on long-term personalization benchmarks demonstrate that LGM significantly outperforms state-of-the-art baselines in capturing both explicit and implicit preferences while enabling personalized responses.
Sources
- A Survey of Large Language Models
- Memory in the Age of AI Agents
- Deep Tabular Research via Continual Experience-Driven Execution
- Toward Native Multimodal Modeling: A Roadmap
- MemoChat: Tuning LLMs to Use Memos for Consistent Long-Range Open-Domain Conversation
- Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
- Reasoning Beyond Language: A Comprehensive Survey on Latent Chain-of-Thought Reasoning
- PersonaMem-v2: Towards Personalized Intelligence via Learning Implicit User Personas and Agentic Memory
- CLR-Bench: Evaluating Large Language Models in College-level Reasoning
- From Local to Global: A Graph RAG Approach to Query-Focused Summarization
- MemGPT: Towards LLMs as Operating Systems
- Mem-{\alpha}: Learning Memory Construction via Reinforcement Learning
- Qwen2.5 Technical Report
- Gemma 3 Technical Report
- Qwen3 Technical Report
- Decoupled Weight Decay Regularization
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection