List your company

Interactive Service Tasks on Tau-bench OOD transfer

41.4Average ScoreSelSkill
31.5234.08536.6539.215May 30, 2026

Evaluation results

MethodLinks
SelSkill2026.0541.43448.75049.2
+Skill2026.0539.63445.241.747.8
Base2026.0531.92241.7——