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

Long-form Question Answering on LongFact MixP (50% polluted)

90.33VeriScore F1@k

MIRAGE

55.011664.180873.3582.5192Jul 6, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.07
90.33
2026.07
87.56
2026.07
86.9
2026.07
84.61
2026.07
81.14
2026.07
80.71
2026.07
73.41
2026.07
73.32
2026.07
73.08
2026.07
71.48
2026.07
70.5
2026.07
70.27
2026.07
69.21
2026.07
67.22
2026.07
66.22
2026.07
65.66
2026.07
65.48
2026.07
65.38
2026.07
65.32
2026.07
65.15
2026.07
64.92
2026.07
61.13
2026.07
59.01
2026.07
58.24
2026.07
57.13
2026.07
56.37