Ranking on Paper Review Benchmark
0.6626Spearman CorrelationGraphReview
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| GraphReview2026.05 | 0.6626 | 0.4808 | 0.8547 | 0.7893 | |
| Gemini-3-Flash-Preview2026.05 | 0.5128 | 0.4181 | 0.8136 | 0.6289 | |
| Qwen3.6-Plus2026.05 | 0.4575 | 0.3708 | 0.639 | 0.5766 | |
| DeepSeek-V4-Pro2026.05 | 0.4046 | 0.3213 | 0.6829 | 0.5707 | |
| GLM-52026.05 | 0.4018 | 0.3274 | 0.8144 | 0.5777 |