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

Question Answering on OBQA (Accuracy, Avg, ΔScore)

88.8Accuracy

ETN

67.864873.299978.73584.1701Mar 13, 2026Mar 17, 2026Mar 22, 2026Mar 26, 2026Mar 31, 2026Apr 4, 2026Apr 9, 2026
Updated 6d ago

Evaluation Results

MethodLinks
2026.04
88.897.15-94.57
2026.04
87.295.3-93.46
2026.04
87.0796.94--
2026.04
86.497.18--
2026.04
86.0751.27--
2026.04
83.4794.51--
2026.03
83.1179.120-
2026.03
81.7877.02-2.1-
2026.04
81.694.33-83.1
2026.04
80.2785.62-76.69
2026.03
78.4475.67-3.45-
2026.03
7470.470-
2026.03
7068.1-2.37-
2026.03
68.6762.72-7.75-