AI Peer Review on PaperAudit-Dataset ICML branch
8.7NoveltyBaseline-Review
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Baseline-ReviewReviewer=Gemini 2.5-Pro, Synth=Origin2026.01 | 8.7 | 8.3 | 8.09 | 8.2 | |
| Baseline-ReviewReviewer=Gemini 2.5-Pro, Synth=GPT-52026.01 | 8.7 | 7.4 | 6.96 | 7.78 | |
| Baseline-ReviewReviewer=Gemini 2.5-Pro, Synth=O4-Mini2026.01 | 8.52 | 7.43 | 6.96 | 7.48 | |
| PaperAudit-ReviewReviewer=Gemini 2.5-Pro, Synth=Origin2026.01 | 8.51 | 6.31 | 6.51 | 6.81 | |
| PaperAudit-ReviewReviewer=Gemini 2.5-Pro, Synth=GPT-52026.01 | 8.48 | 5.55 | 5.45 | 6.31 | |
| PaperAudit-ReviewReviewer=Gemini 2.5-Pro, Synth=O4-Mini2026.01 | 8.43 | 5.73 | 5.69 | 6.07 | |
| DeepReviewReviewer=Gemini 2.5-Pro, Synth=Origin2026.01 | 7.9 | 7.68 | 7.75 | 7.53 | |
| DeepReviewReviewer=Gemini 2.5-Pro, Synth=O4-Mini2026.01 | 7.74 | 6.7 | 6.61 | 6.57 | |
| DeepReviewReviewer=Gemini 2.5-Pro, Synth=GPT-52026.01 | 7.71 | 6.63 | 7.21 | 6.82 | |
| Baseline-ReviewReviewer=GPT-5, Synth=Origin2026.01 | 7.61 | 6.63 | 7.19 | 7.27 | |
| PaperAudit-ReviewReviewer=GPT-5, Synth=Origin2026.01 | 7.61 | 6.15 | 6.55 | 6.93 | |
| DeepReviewReviewer=GPT-5, Synth=Origin2026.01 | 7.59 | 6.91 | 7.64 | 7.35 | |
| DeepReviewReviewer=GPT-5, Synth=GPT-52026.01 | 7.39 | 6.07 | 6.74 | 6.65 | |
| Baseline-ReviewReviewer=GPT-5, Synth=GPT-52026.01 | 7.3 | 5.75 | 6.17 | 6.63 | |
| PaperAudit-ReviewReviewer=GPT-5, Synth=GPT-52026.01 | 7.29 | 5.45 | 5.89 | 6.45 | |
| DeepReviewReviewer=GPT-5, Synth=O4-Mini2026.01 | 7.12 | 6.07 | 6.55 | 6.54 | |
| PaperAudit-ReviewReviewer=GPT-5, Synth=O4-Mini2026.01 | 6.99 | 5.55 | 5.76 | 6.28 | |
| Baseline-ReviewReviewer=GPT-5, Synth=O4-Mini2026.01 | 6.97 | 5.79 | 6.09 | 6.41 |