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

Natural Language Processing Reasoning on BBH-NLP (test)

66.2Accuracy (ACC)

KP

Updated 1d ago

Evaluation Results

Method	Links
KP 2025.09		66.2	-	100	0
KAPPA 2025.09		63.6	-	74.9	0.87
KP 2025.09		63.5	-	100	0
KAPPA 2025.09		63	-	74	0.89
KAPPA 2025.09		63	-	83.7	0.39
KAPPA 2025.09		62.8	-	73.2	0.89
KAPPA 2025.09		62.4	-	86.3	0.47
CAA 2025.09		61.2	-	69.9	0.92
Qwen 2.5 7B Instruct 2025.09		61.1	5.1	69.8	0.92
Base 2025.09		61.1	-	69.8	0.92
DoLA 2025.09		61	-	69.3	0.92
CAA 2025.09		60.2	-	73.3	0.41
KAPPA 2025.09		60.2	-	75.3	0.46
Llama 3.1 8B Instruct 2025.09		59.6	3.9	73.9	0.41
Base 2025.09		59.6	-	73.9	0.41
DoLA 2025.09		56.6	-	69.6	0.51