Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Tool-using on NoisyToolBench IwE

1.3Average Steps

CoT

-27.6536167.7832363.22558.6568Aug 31, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.08
1.3-
2024.08
20.12
2024.08
2.16-
2024.08
2.48-
2024.08
2.72-
2024.08
2.86-
2024.08
2.88-
2024.08
2.920.04
2024.08
3.04-
2024.08
3.12-
2024.08
3.180.46
2024.08
3.34-
2024.08
3.460.36
2024.08
3.660.44
2024.08
3.660.4
2024.08
3.760.36
2024.08
3.9-
2024.08
4.160.48
2024.08
4.540.48
2024.08
5.080.76
2024.08
5.78-
2024.08
5.98-
2024.08
8.560.88
2024.08
9.420.62
2024.08
11.680.94
2024.08
11.821.2
2024.08
13.081.04
2024.08
13.82-
2024.08
22.4-
2024.08
24.24-
2024.08
411.181.94
2024.08
725.14-