arXiv cs.LG
· Papers
DREvo: Distilling Recalibrated Historical Experience for Harness Self-Evolution
arXiv:2607.26722v2 Announce Type: replace-cross Abstract: Harness plays a critical role in large language model agent performance, and building a high-performing harness requires substantial expert effort. Therefore, recent research has increasingly explored harness self-evolution, which iteratively proposes, evaluates