Benchmark RadarSciDraw-Bench
SciDraw-Bench is a benchmark for scientific figure generation, with 32 tasks across eight figure types and ten disciplines. Each task pairs a natural-language prompt with a machine-checkable specification, and evaluation uses four dimensions: Text Fidelity, Semantic Correctness, Structural Quality, and Convention Adherence.
- Released
- 2026-06-24
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Fills the gap in evaluating scientific figure generation, which requires correct labels and diagrammatic structure beyond natural-image composition. It provides a protocol for measuring usability of generated figures, aiding development of domain-specific generative systems.
Motivation
Text-to-image and multimodal generative models are increasingly used to produce scientific figures such as mechanism diagrams, experimental-design schematics, conceptual frameworks, and graphical abstracts.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.