Diagram Generation Quality Assessment on Diagram Generation Quality Evaluation Dataset
8.2Aesthetic ScoreDiagramRAG
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| DiagramRAGbackbone=GPT2026.05 | 8.2 | 7.105 | 8.2 | 7.3 | 6.5 | 6.4 | 6.502 | 7.1 | 7.12 | |
| DiagramRAGbackbone=NanoBanana2026.05 | 8 | 7.1 | 8 | 7.4 | 6.9 | 6.704 | 6.7 | 7.101 | 7.17 | |
| Autofigure2026.05 | 7.9 | 6.8 | 7.9 | 7.8 | 6.9 | 6.6 | 6.5 | 6.8 | 7.06 | |
| GPT-image-22026.05 | 7.6 | 6.4 | 7.6 | 8 | 7 | 6.7 | 6.4 | 6.4 | 6.92 | |
| Paperbanana2026.05 | 6.625 | 5.625 | 6.625 | 7.125 | 6.375 | 5.75 | 5.625 | 5.5 | 6.188 | |
| NanoBanana2026.05 | 6.5 | 5.5 | 6.5 | 7.4 | 7.4 | 6.6 | 6.1 | 5.5 | 6.5 |