ScienceQA
multimodal
+
+
+
+
About
ScienceQA is the first large-scale multimodal science question-answering benchmark featuring approximately 21,000 multiple-choice questions with diverse science topics, lectures, and explanations. This comprehensive evaluation tests multi-hop reasoning abilities and interpretability across various scientific domains, combining visual and textual information to assess AI models' scientific understanding and explanatory capabilities.
+
+
+
+
Evaluation Stats
Total Models1
Organizations1
Verified Results0
Self-Reported1
+
+
+
+
Benchmark Details
Max Score1
Language
en
+
+
+
+
Performance Overview
Score distribution and top performers
Score Distribution
1 models
Top Score
91.3%
Average Score
91.3%
High Performers (80%+)
1Top Organizations
#1Microsoft
1 model
91.3%
+
+
+
+
Leaderboard
1 models ranked by performance on ScienceQA
| License | Links | ||||
|---|---|---|---|---|---|
| Aug 23, 2024 | MIT | 91.3% |
