Query-to-Presentation Generation on PresentEval Discussion Presentation 1.0 (test)
4.85Quiz ScorePresentAgent-2
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| PresentAgent-2Model=Qwen3.5-VL-Plus2026.05 | 4.85 | 4.43 | 4.22 | 4.47 | 4.37 | |
| Human ReferenceModel=Human-created2026.05 | 4.83 | 4.45 | 4.31 | 4.45 | 4.4 | |
| PresentAgent-2Model=Claude Opus 4.72026.05 | 4.82 | 4.43 | 4.21 | 4.49 | 4.38 | |
| PresentAgent-2Model=Gemini 3.1 Pro2026.05 | 4.8 | 4.37 | 4.08 | 4.3 | 4.25 | |
| PresentAgent-2Model=GPT-5.52026.05 | 4.77 | 4.19 | 4.11 | 4.2 | 4.17 | |
| PresentAgent-2Model=GLM-4.7V2026.05 | 4.67 | 4.2 | 4.03 | 4.11 | 4.11 |