GEPA
Refelctive Prompt Optimisation can Outperform Reinforcement Learning
ALHF
Agent Learning through Human Feedback (DataBricks)
AIME Adaptable Multi Agent Systems
A practical implementation of Multi-Agent Systems (MAS)
Why Do Multi-Agent LLM Systems Fail?
MAST (Multi-Agent System Failure Taxonomy)
Self Improvement
A comprehensive look at how AI systems are learning to improve themselves