GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning

4 pointsposted 10 hours ago
by handfuloflight
(arxiv.org)

No comments yet