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

Fact-checking on DeepFact-Bench (test)

87.2Accuracy

DeepFact-Eval

51.11260.48169.8579.219Mar 6, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.03
87.289.987.991.9
2026.03
83.486.985.788.2
2026.03
81.58584.685.3
2026.03
77.983.178.588.2
2026.03
76.382.276.489
2026.03
72.577.478.176.6
2026.03
69.179.766.798.9
2026.03
68.869.55886.7
2026.03
68.379.366.199.2
2026.03
58.563.269.258.3
2026.03
55.95376.340.6
2026.03
5558.367.751.2
2026.03
52.548.971.937