List your company

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

40AdherenceVanilla
-1.69.22030.8Jun 24, 2026

Evaluation results

MethodLinks
VanillaBackbone=Qwen2.5-1.5B-Instruct2026.064032522.531.6
SFT_BFCLBackbone=Qwen2.5-1.5B-Instruct2026.063020442.524.1
ETSBackbone=Qwen2.5-1.5B-Instruct2026.062622362.521.6
PRSBackbone=Qwen2.5-1.5B-Instruct2026.062630434.225.8
OPSBackbone=Qwen2.5-1.5B-Instruct2026.062016506.723.2
SFT_BFCL + RLBackbone=Qwen2.5-1.5B-Instruct2026.06881006.5
GRPOBackbone=Qwen2.5-1.5B-Instruct2026.0600000
HBGBackbone=Qwen2.5-1.5B-Instruct2026.0600000