awesome-personalized-ai

A living, evidence-based tracker of personalized LLM & agent research — benchmarks, methods, and surveys across preference alignment, long-term memory, multimodal, tool use, safety and privacy.

Personalization is not one score. A system may remember a fact yet apply it at the wrong time, follow a preference while degrading factuality, or complete a task while leaking private information. This repository tracks the benchmarks, datasets, protocols, methods, surveys, position papers, and analyses that define personalized LLM and agent research—with the evaluation resources needed to test the full personalization lifecycle kept front and center.