AI RESEARCH
FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast
arXiv CS.AI
•
ArXi:2605.16233v1 Announce Type: new Can LLM agents improve decision-making through self-generated memory without gradient updates? We propose FORGE (Failure-Optimized Reflective Graduation and Evolution), a staged, population-based protocol that evolves prompt-injected natural-language memory for hierarchical ReAct agents.