DisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing
Recent image generation and editing models can produce visually appealing natural images, yet they remain unreliable when the target image is a knowledge-intensive diagram whose correctness depends on disciplinary concepts, symbolic structure, and precise spatial relations. We introduce DisciplineGen-1M, a million-scale multidisciplinary dataset that supports text-to-image generation and image editing. It contains 1.2M samples spanning mathematics, physics, chemistry, biology, geography, computer science, economics, history, music, and sports. To construct the dataset, we design a scalable fra
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 47%voxel51/fiftyone →
- LinkedLinked via arxiv author · 85%Zhaokai Wang →
“DisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing”
- LinkedLinked via arxiv author · 85%Mingxin Liu →
“DisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing”
- LinkedLinked via arxiv author · 85%Zirun Zhu →
“DisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing”
- LinkedLinked via arxiv author · 85%Ziqian Fan →
“DisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing”
- LinkedLinked via arxiv author · 85%Yiguo He →
“DisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing”
- LinkedLinked via arxiv author · 85%Mohan Zhang →
“DisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing”
- LinkedLinked via arxiv author · 85%Leyao Gu →
“DisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing”