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

Tool-using on NoisyToolBench IBTC

1Average Steps

CoT

-21.3512129.5194280.39431.2606Aug 31, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.08
1-
2024.08
1.020.02
2024.08
1.020
2024.08
1.04-
2024.08
1.08-
2024.08
1.10
2024.08
1.10.02
2024.08
1.140
2024.08
1.160.04
2024.08
1.16-
2024.08
1.18-
2024.08
1.28-
2024.08
1.320.04
2024.08
1.460.46
2024.08
1.460.46
2024.08
1.540.3
2024.08
1.540.24
2024.08
1.680.1
2024.08
1.9-
2024.08
2.04-
2024.08
2.1-
2024.08
2.10.16
2024.08
2.41.1
2024.08
2.760.14
2024.08
3.941.7
2024.08
4.06-
2024.08
4.26-
2024.08
4.52-
2024.08
5.5-
2024.08
8.98-
2024.08
11.64-
2024.08
559.78-