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

Question Answering on PubMedQA (Accuracy)

83.67Accuracy

TL++ secure

50.3959.0367.6776.31Jun 15, 2026Jun 16, 2026Jun 18, 2026Jun 19, 2026Jun 21, 2026Jun 22, 2026Jun 24, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
83.67
2026.06
82.6
2026.06
81.33
2026.06
81.2
2026.06
81.07
2026.06
80.87
2026.06
76.6
2026.06
75
2026.06
74.6
2026.06
73.2
2026.06
72.8
2026.06
72.8
2026.06
72.8
2026.06
72
2026.06
71.8
2026.06
70.8
2026.06
68.8
2026.06
68.8
2026.06
67.44
2026.06
67.11
2026.06
65.67
2026.06
65.4
2026.06
65.33
2026.06
64
2026.06
62.2
2026.06
61.44
2026.06
59.6
2026.06
59.44
2026.06
57.4
2026.06
55.89
2026.06
55.8
2026.06
55.6
2026.06
55.6
2026.06
55.44
2026.06
55.4
2026.06
55.4
2026.06
55.2
2026.06
55.2
2026.06
53.67
2026.06
51.67