AI RESEARCH
Self-Evolving LLM Memory Extraction Across Heterogeneous Tasks
arXiv CS.CL
•
ArXi:2604.11610v1 Announce Type: new As LLM-based assistants become persistent and personalized, they must extract and retain useful information from past conversations as memory. However, the types of information worth remembering vary considerably across tasks. We formalize the \textit{heterogeneous memory extraction} task and