List your company

Tool-use Generalization on ACEBench Format and Content OOD v1 (test)

38Adherence (Ad.)Vanilla
-1.528.741929.26Jun 24, 2026

Evaluation results

MethodLinks
VanillaBackbone=Qwen2.5-1.5B-Instruct2026.06382046026
OPSBackbone=Qwen2.5-1.5B-Instruct2026.06362443025.75
GRPOBackbone=Qwen2.5-1.5B-Instruct2026.06322443024.75
HBGBackbone=Qwen2.5-1.5B-Instruct2026.06282042022.5
PRSBackbone=Qwen2.5-1.5B-Instruct2026.0616824012
ETSBackbone=Qwen2.5-1.5B-Instruct2026.061002508.75
SFT_BFCLBackbone=Qwen2.5-1.5B-Instruct2026.0600000
SFT_BFCL + RLBackbone=Qwen2.5-1.5B-Instruct2026.0600000