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

Science Question Answering on ScienceQA (SQA)

97.4Accuracy

GPT-5-high

69.00876.37983.7591.121Dec 5, 2024Mar 10, 2025Jun 13, 2025Sep 17, 2025Dec 21, 2025Mar 26, 2026Jun 30, 2026
Updated 24d ago

Evaluation Results

MethodLinks
2026.06
97.4-
2026.06
96.2-
2026.06
95-
2026.06
92.8-
2026.06
92.76-
2026.06
91.96-
2026.06
91.8-
2026.06
91.58-
2026.06
90.8-
2026.06
90.2-
2026.06
89.5-
2026.02
88.8-
2026.02
88.7-
2026.02
88.65-
2026.06
88.4-
2026.02
88-
2026.06
87.8-
2026.06
87.6-
2026.03
87.33-
2026.06
87-
2026.06
85.3-
2026.02
84.88-
2026.02
84.8-
2026.03
84.28-
2026.02
84.2-
2026.03
84.13-
2026.03
82.15-
2026.02
82-
2026.04
79.7-
2025.06
79.4-
2026.04
79.3-
2026.03
78.45-
2026.03
78.29-
2025.06
76.7-
2026.02
76.7-
2026.02
76.5-
2025.06
76.5-
2026.03
76.5-
2026.02
76.3-
2026.02
76.3-
2026.06
75.4-
2026.06
75.1-
2026.06
75-
2026.06
75-
2026.02
74.9-
2026.06
74.9-
2026.06
74.9-
2026.02
74.8-
2026.06
74.7-
2024.12
74.4-
2026.06
74.2-
2026.06
74.2-
2026.02
74.1-
2026.02
74.1-
2024.12
74-
2026.03
73.97-
2026.06
73.6-
2024.12
73.5-
2026.04
73.5-
2026.06
73.1-
2026.03
73.08-
2026.03
73.08-
2024.12
73-
2026.03
73-
2026.03
72.89-
2026.03
72.89-
2024.12
72.8-
2026.03
72.78-
2024.12
72.7-
2024.12
72.3-
2026.03
72.2-
2025.11
72.2-
2025.11
72.1-
2026.03
71.89-
2026.02
71.6-
2026.03
71.6-
2026.03
71.6-
2026.03
71.6-
2025.11
71.6-
2026.03
71.5-
2026.02
71.4-
2026.03
71.3-
2026.03
71.1-
2024.12
71100.4
2026.02
70.9-
2025.11
70.9-
2026.03
70.82-
2024.12
70.7100
2024.12
70.7100
2026.03
70.64-
2025.12
70.6-
2025.12
70.4-
2026.03
70.32-
2025.09
70.2-
2024.12
70.299.3
2026.04
70.2-
2026.02
70.1-
2026.02
70.1-
2024.12
70.199.2
2024.12
70.199.2
Showing 100 of 338 rows