List your company

Agent Skill Acquisition on tau2-bench

81.2Pass@1Qwen3.5 35B-A3B
47.71256.40665.173.794Feb 18, 2026Feb 22, 2026Feb 27, 2026Mar 4, 2026Mar 9, 2026Mar 14, 2026Mar 19, 2026

Evaluation results

MethodLinks
Qwen3.5 35B-A3B2026.0381.2——
ECScontext_strategy=Evolutionary Context Search2026.0276.77168.3
Fullcontext_strategy=Full context concatenation2026.0271.762.855
Randomcontext_strategy=Random chunk selection2026.026555.650
BM25context_strategy=BM25 retrieval2026.0264.15548.3
Plain LLMcontext_strategy=None (No external context)2026.0262.248.940
Nemotron-3-Super 120B-A12B2026.0361.2——
Nemotron-Cascade-2 30B-A3B2026.0358.9——
Nemotron-3-Nano 30B-A3B2026.0349——