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

Science Question Answering on ScienceQA (EM)

88.45Exact Match (EM)

Baseline

65.299671.309877.3283.3302Feb 10, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
88.45
2026.02
85.52
2026.02
85.23
2026.02
85.08
2026.02
84.33
2026.02
84.18
2026.02
84.18
2026.02
83.99
2026.02
83.94
2026.02
83.49
2026.02
83.24
2026.02
82.8
2026.02
82.65
2026.02
82.25
2026.02
81.95
2026.02
81.11
2026.02
80.96
2026.02
80.96
2026.02
79.28
2026.02
66.88
2026.02
66.73
2026.02
66.68
2026.02
66.19