List your company

Agentic Tool-Use on tau-bench Retail and Airline

78.3Retail Success RateModSFT+RL
69.9872.1474.376.46Jun 7, 2026

Evaluation results

MethodLinks
ModSFT+RLBackbone=Qwen3-8B2026.0678.316.147.2
Rejuvenation+RLBackbone=Qwen3-8B2026.067717.547.3
DFT+RLBackbone=Qwen3-8B2026.0675.312.243.8
OverSFT+RLBackbone=Qwen3-8B2026.0670.319.945.1