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

Knowledge Acquisition on TOFU author-profile questions (held-out)

84.3Task Accuracy

FINCH

49.4658.50567.5576.595May 19, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.05
84.3
2026.05
83.5
2026.05
83.5
2026.05
83.3
2026.05
83.3
2026.05
82.8
2026.05
82.5
2026.05
82.3
2026.05
82
2026.05
80.8
2026.05
77.5
2026.05
76
2026.05
69.5
2026.05
65.8
2026.05
63.3
2026.05
62.5
2026.05
58.8
2026.05
55.8
2026.05
53.8
2026.05
53.5
2026.05
52.8
2026.05
50.8