1,650-question STEM diagram dataset collected via campus-wide submission campaign

Product or Programme · All products and programmes
Evidence · 1 claim (1 for) · 1 study (1 design), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description

A restricted dataset of 1,650 educator- and student-submitted STEM questions containing structurally meaningful diagrams for evaluating AI interpretation of topological reasoning, collected and curated by Sun et al.

Components

Claims

• StructRAG achieves 89.3% question-level accuracy and 93.0% edge-level F1 on 1,650 STEM diagram questions, outperforming OCR+CV, GPT-4 graph-only, direct-image GPT-4o, and ablated variants [+M]

Related Products and Programmes

• Curated library of 375 abstract graph templates built with 12 STEM educators
• Multimodal OBE examination-answer dataset of 485 scanned submissions with criterion-level rubric marks

Key Sources

• Sun Y., Liao Y. & Ma X. (2026). Advancing diagram-based reasoning in AI tutoring systems: a structural approach for STEM education. Smart Learning Environments. https://doi.org/10.1186/s40561-026-00459-9

Page metadata

Field	Value
type	product
id	1650-question-stem-diagram-dataset-collected-via-campus-wide-submission-campaign
title	1,650-question STEM diagram dataset collected via campus-wide submission campaign
description	A restricted dataset of 1,650 educator- and student-submitted STEM questions containing structurally meaningful diagrams for evaluating AI interpretation of topological reasoning, collected and curated by Sun et al.
status	review
product_kind	dataset
generated	by process:settle-candidates on 2026-10-11
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.