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

MT-Bench Real-data 20% verified Evaluation

77.82Original Accuracy

Logistic Regression

73.92975.874577.8279.7655Jun 18, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
77.8275.39-2.43513
2026.06
77.8277.66-0.16513
2026.06
77.8277.28-0.54513
2026.06
77.8276.74-1.07513
2026.06
77.8277.830.01513
2026.06
77.8281.013.1987.4