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

Function Selection on TaskBench HuggingFace

77.1Function Selection Accuracy

Raw Demonstrations

Updated 1mo ago

Evaluation Results

Method	Links
Raw Demonstrations 2025.12		77.1
Raw Demonstrations 2025.12		76.7
No ICL 2025.12		74.8
No ICL 2025.12		74.4
Raw Demonstrations 2025.12		72.5
DTDR-C 2025.12		71.9
DTDR-L 2025.12		71.8
DTDR-C 2025.12		71.5
DTDR-L 2025.12		71.2
No ICL 2025.12		70.5
Dependency Retrieval 2025.12		70
Dependency Retrieval 2025.12		69.2
Dependency Retrieval 2025.12		67.3
DTDR-C 2025.12		67.3
DTDR-L 2025.12		63.8
DTDR-L 2025.12		60.5
Learned Retrieval 2025.12		56.9
DTDR-C 2025.12		56
Learned Retrieval 2025.12		55.2
Learned Retrieval 2025.12		53.5
DTDR-C 2025.12		52.8
Dependency Retrieval 2025.12		52.8
Learned Retrieval 2025.12		52
Raw Demonstrations 2025.12		49.8
DTDR-L 2025.12		49
No ICL 2025.12		45.8
Raw Demonstrations 2025.12		43.1
Learned Retrieval 2025.12		42.8
Dependency Retrieval 2025.12		41.6
No ICL 2025.12		31.2
Dynamic LR (DTDR-L) 2025.12		27.8
Dynamic DR (DTDR-C) 2025.12		27.5
DTDR-C 2025.12		20.8
LR (Erdogan et al. 2024) 2025.12		20.5
DTDR-L 2025.12		18.9
DR (Liu et al. 2024a) 2025.12		17.1
QTS (Gao et al. 2025) 2025.12		14.6
QTS (Paramanayakam et al. 2025) 2025.12		12.9
Learned Retrieval 2025.12		10
BM-25 2025.12		8.3
QTS (Vanilla) 2025.12		7.9
Dependency Retrieval 2025.12		6.7
No ICL 2025.12		5.5
Raw Demonstrations 2025.12		4.3
Random Guess 2025.12		4.2