A 201-question multimodal STEM dataset with student records shows that the best model (58.5%) trails students (62.7%), with the largest gaps on essential-image and multi-concept questions.
MCQ" (Multiple Choice Question) –
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CY 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Challenges for AI in Multimodal STEM Assessments: a Human-AI Comparison
A 201-question multimodal STEM dataset with student records shows that the best model (58.5%) trails students (62.7%), with the largest gaps on essential-image and multi-concept questions.