List your company

Tool Use on Evaluation dataset

51.98AccuracyPORTool
10.463221.241632.0242.7984Oct 29, 2025

Evaluation results

MethodLinks
PORToolBackbone=Qwen-2.5-7B-Instruct2025.1051.983.077.10.851
AgentFlowBackbone=Qwen-2.5-7B-Instruct2025.1048.233.311.210.816
ARPOBackbone=Qwen-2.5-7B-Instruct2025.1048.183.1912.340.819
SimpleTIRBackbone=Qwen-2.5-7B-Instruct2025.1047.623.1210.470.832
GRPOfmBackbone=Qwen-2.5-7B-Instruct2025.1047.583.1811.520.826
PORToolBackbone=Qwen-3-1.7B2025.1046.62.647.290.877
GRPOBackbone=Qwen-2.5-7B-Instruct2025.1046.093.3113.270.808
ToRLBackbone=Qwen-2.5-7B-Instruct2025.1045.523.212.330.814
ARPOfmBackbone=Qwen-2.5-7B-Instruct2025.1044.973.1611.560.827
DAPOBackbone=Qwen-2.5-7B-Instruct2025.1043.513.3714.580.79
GRPOfmBackbone=Qwen-3-1.7B2025.1042.762.799.910.857
SimpleTIRBackbone=Qwen-3-1.7B2025.1041.762.8211.030.884
ToRLBackbone=Qwen-3-1.7B2025.1041.732.9411.40.81
DAPOBackbone=Qwen-3-1.7B2025.1039.473.0811.960.793
ARPOBackbone=Qwen-3-1.7B2025.1039.273.0711.960.784
GRPOBackbone=Qwen-3-1.7B2025.1039.082.9610.650.834
AgentFlowBackbone=Qwen-3-1.7B2025.1037.23.2113.270.767
ARPOfmBackbone=Qwen-3-1.7B2025.1034.563.2111.030.808
PromptingBackbone=Qwen-2.5-7B-Instruct2025.1024.364.7658.50.466
PromptingBackbone=Qwen-3-1.7B2025.1012.065.583.920.322