Memory Has Geometry: Non-Uniform Geometric Memory for Long-Horizon Personalized AI
cs.AI
Submitted: 2026-09-16
Updated: 2026-09-16
Comments: Accepted to the IEEE ICDM 2026 (BlueSky). 6 pages, 2 figures, and 2 tables
License: http://creativecommons.org/licenses/by/4.0/
The gist: Long-term memory is becoming a core substrate for personalized AI, yet most systems still represent personalization as discrete records in a largely static latent space, accessed under one global
Terminology
Abstract
Long-term memory is becoming a core substrate for personalized AI, yet most systems still represent personalization as discrete records in a largely static latent space, accessed under one global similarity notion. For data mining, this creates a mismatch: the evidence is a temporal event stream, while the dominant abstraction is a searchable record set. We argue that long-horizon personalization should instead model memory as a user-specific dynamical state space with locally heterogeneous geometry. Geometry here is a computational language, not a literal claim about cognition: it captures stable versus volatile regions, variable-rate drift, heterogeneous neighborhoods, and uncertainty about current user state. Profiles and isolated events remain useful as points, but interaction, feedback, and elapsed time induce trajectories. Memory access then becomes trajectory-conditioned reconstruction of the relevant user state, not only nearest-neighbor lookup.
Sources
- User-LLM: Efficient LLM Contextualization with User Embeddings
- A Survey of Personalized Large Language Models: Progress and Future Directions
- A Survey on the Memory Mechanism of Large Language Model based Agents
- MemoryBank: Enhancing Large Language Models with Long-Term Memory
- Zep: A Temporal Knowledge Graph Architecture for Agent Memory
- SemaCDR: LLM-Powered Transferable Semantics for Cross-Domain Sequential Recommendation
- Generative Archetype-Grounded Item Representations for Sequential Recommendation
- Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
- Position: Beyond Euclidean -- Foundation Models Should Embrace Non-Euclidean Geometries
- Temporal Graph Networks for Deep Learning on Dynamic Graphs
- Enhancing Hyperbolic Graph Embeddings via Contrastive Learning
- Hyperbolic Multimodal Continual Learning
- ProWorld: Progress-Aware Hyperbolic World Models for Long-Horizon Visual Goal Reaching
- Low-Rank Adaptation for Foundation Models: A Comprehensive Review
- Towards trustworthy agentic AI: a comprehensive survey of safety, robustness, privacy, and system security
- LifeBench: A Benchmark for Long-Horizon Multi-Source Memory
- TRACE: Trajectory-Aware Comprehensive Evaluation for Deep Research Agents
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