Jialin Zhang(@benzhang04):1/ We just made GEPA’s optimization loop much faster, along with improved generalization! Instead of proposing and evaluating one candidate at a time, each GEPA step now proposes a batch of candidates and evaluates them all concurrently. Parallel proposals significantly cut wall-clock time (3–4× faster) while, surprisingly, also generalizing better (up to +11 pts). 🧵

Jialin Zhang

@benzhang04

CS @ UC Berkeley | Undergraduate researcher @berkeley_ai & @BerkeleySky

加入 June 2025

30 正在關注 77 粉絲

Jialin Zhang@benzhang04

2026.08.05 17:57

1/ We just made GEPA’s optimization loop much faster, along with improved generalization! Instead of proposing and evaluating one candidate at a time, each GEPA step now proposes a batch of candidates and evaluates them all concurrently. Parallel proposals significantly cut wall-clock time (3–4× faster) while, surprisingly, also generalizing better (up to +11 pts). 🧵