STRING on ShellOps benchmark leaderboard · SOTA2 Research

STRING on ShellOps

49.1LLM Judge AccuracyA3

Evaluation Results

Method	Links
A32026.05		49.1
A32026.05		49
LATS2026.05		28.8
ReACT2026.05		28.3
GSPO2026.05		27.7
GiGPO2026.05		27
HGPO2026.05		23.9
RetroAgent2026.05		20
rStar2026.05		18.6