Novelty Report Generation on 50-paper
9.67CompletenessNoveltyAgent
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| NoveltyAgentSystem Category=AI Reviewer Systems2026.03 | 9.67 | 9.55 | 9.09 | 8.4 | 9.93 | 9.33 | |
| GPT-5 DeepResearchSystem Category=Web Research Systems2026.03 | 8.34 | 8.4 | 9.2 | 6.75 | 9.66 | 8.47 | |
| GPT-5 ThinkingSystem Category=Web Research Systems2026.03 | 8.32 | 8.01 | 9.18 | 5.89 | 10 | 8.28 | |
| AgentReviewerSystem Category=AI Reviewer Systems2026.03 | 7.59 | 9.14 | 9.23 | 6.43 | 9.82 | 8.44 | |
| DeepReviewSystem Category=AI Reviewer Systems2026.03 | 7.41 | 9.09 | 9.25 | 7.54 | 9.42 | 8.54 | |
| Gemini-2.5-Flash DeepResearchSystem Category=Web Research Systems2026.03 | 6.88 | 8.56 | 9.21 | 5.74 | 8.3 | 7.74 | |
| Kimi-2System Category=Web Research Systems2026.03 | 6.23 | 7.63 | 9.21 | 5.5 | 9.96 | 7.71 |