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

Tool Learning under Instruction with Error on NoisyToolBench IwE 1.0 (test)

74A1 Success Rate

CoT + AwN

-2.9617.023756.98Aug 31, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.08
744828
2024.08
724236
2024.08
663626
2024.08
644638
2024.08
604432
2024.08
582822
2024.08
584232
2024.08
583836
2024.08
541210
2024.08
543836
2024.08
543822
2024.08
522624
2024.08
523836
2024.08
502416
2024.08
503422
2024.08
481414
2024.08
483434
2024.08
403026
2024.08
3662
2024.08
363630
2024.08
343426
2024.08
221212
2024.08
1864
2024.08
121210
2024.08
122018
2024.08
101414
2024.08
8126
2024.08
82624
2024.08
044
2024.08
062
2024.08
086
2024.08
084