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

Question Answering on TriviaQA LongBench

91.7F1 Score

InnerQHybrid

37.422451.513765.60579.6963Feb 26, 2026
Updated 12d ago

Evaluation Results

MethodLinks
2026.02
91.7
2026.02
91.7
2026.02
91.64
2026.02
91.51
2026.02
91.33
2026.02
90.53
2026.02
88.27
2026.02
87.97
2026.02
87.85
2026.02
87.83
2026.02
87.47
2026.02
84.55
2026.02
83.84
2026.02
83.51
2026.02
83.32
2026.02
82.77
2026.02
82.27
2026.02
69.39
2026.02
67.53
2026.02
67.48
2026.02
67.29
2026.02
65.41
2026.02
65.37
2026.02
64.75
2026.02
64.03
2026.02
63.92
2026.02
62.92
2026.02
59.2
2026.02
59.06
2026.02
39.51