When Reviews Disagree: leaderboard
Scientific peer reviews frequently contain conflicting expert judgments, and the increasing scale of conference submissions makes it challenging for Area Chairs and editors to reliably identify and interpret such disagreements. Existing approaches typically frame reviewer disagreement as binary contradiction detection over isolated sentence pairs, abstracting away the review-level context and obsc
Metric: ρ (self-reported). Source: benchmarklist.com. Saturation forecast: Around March 2027. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash | 45.23 | #99 |
| 2 | GPT-5.2 | 36.79 | #110 |
| 3 | Llama 3 8B Instruct | 35.41 | #1129 |
| 4 | Llama 4 Maverick | 33.15 | #464 |
| 5 | Qwen 2.5 7B Instruct | 28.54 | #862 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=when-reviews-disagree · How It Works · Data refreshed daily, snapshot 2026-10-20.