bowen-upenn/PersonaMem
197 12PythonMITUpdated 2 months ago
[COLM 2025] Know Me, Respond to Me: Benchmarking LLMs for Dynamic User Profiling and Personalized Responses at Scale
benchmarkchatbotlarge-language-modelsllm-memorylong-contextpersonapersonalizationpersonalized-generation
Our verdict: worth borrowing ideas from
Contains a full pipeline and evaluation code for dynamic user profiling that can be harvested for persona‑state‑machine logic even if the repo itself isn’t directly used