{"slug": "memory-has-geometry-non-uniform-geometric-memory-for-long-horizon-personalized", "title": "Memory Has Geometry: Non-Uniform Geometric Memory for Long-Horizon Personalized AI", "summary": "A new arXiv paper (2609.17969v1) argues that long-horizon personalized AI should model memory as a user-specific dynamical state space with locally heterogeneous geometry rather than as discrete records in a static latent space accessed by a single global similarity notion. The authors state that profiles and isolated events remain useful as points, but interaction, feedback, and elapsed time induce trajectories, making memory access a trajectory-conditioned reconstruction of the relevant user state instead of only nearest-neighbor lookup. The paper frames geometry as a computational language, not a literal claim about cognition, capturing stable versus volatile regions, variable-rate drift, heterogeneous neighborhoods, and uncertainty about current user state.", "body_md": "arXiv:2609.17969v1 Announce Type: new \nAbstract: 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.", "url": "https://wpnews.pro/news/memory-has-geometry-non-uniform-geometric-memory-for-long-horizon-personalized", "canonical_source": "https://www.machinebrief.com/news/memory-has-geometry-non-uniform-geometric-memory-for-long-ho-797v", "published_at": "2026-09-17 04:00:00+00:00", "updated_at": "2026-09-17 04:55:03.143349+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "ai-research", "large-language-models"], "entities": ["arXiv"], "alternates": {"html": "https://wpnews.pro/news/memory-has-geometry-non-uniform-geometric-memory-for-long-horizon-personalized", "markdown": "https://wpnews.pro/news/memory-has-geometry-non-uniform-geometric-memory-for-long-horizon-personalized.md", "text": "https://wpnews.pro/news/memory-has-geometry-non-uniform-geometric-memory-for-long-horizon-personalized.txt", "jsonld": "https://wpnews.pro/news/memory-has-geometry-non-uniform-geometric-memory-for-long-horizon-personalized.jsonld"}}