Beyond Rating - Rating MAE: leaderboard
Beyond Rating (2026) evaluates AI peer reviewers on the text of their reviews: generated reviews of ICLR and NeurIPS papers are split into atomic points and matched by Qwen3-235B against filtered high-confidence human reviews. This board is the mean absolute error of the review's overall rating against the human ratings on the ICLR scale; lower is better.
Metric: Mean absolute error between the review's overall rating and the human reviewers' ratings on the ICLR scale (1, 3, 5, 6, 8, 10), on 1,000 test papers from ICLR 2024-2026 and NeurIPS 2022-2025 with three to five high-confidence human reviews each; lower is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 15 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepReviewer-14B | 0.85 | #844 |
| 2 | OpenReviewer-8B | 0.86 | |
| 3 | DeepSeek V3.2 | 0.98 | #419 |
| 4 | Qwen 3 235B A22B 2507 Instruct | 0.98 | #616 |
| 5 | Qwen 3 8B | 1.03 | #1873 |
| 6 | SEA-E-7B | 1.1 | |
| 7 | Claude Sonnet 4.5 | 1.14 | #261 |
| 8 | GPT-5.2 | 1.14 | #185 |
| 9 | AgentReview + GPT-5 | 1.28 | |
| 10 | CycleReviewer-70B | 1.29 | #1880 |
| 11 | Qwen 3 30B A3B 2507 Instruct | 1.41 | #1100 |
| 12 | Gemini 3 Pro (Preview) | 1.52 | #119 |
| 13 | Llama 3.1 70B Instruct | 2.32 | #1571 |
| 14 | Llama 3.1 8B Instruct | 2.51 | #4600 |
| 15 | AI Scientist + GPT-5 | 2.68 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=beyond-rating-rating-mae · How It Works · Data refreshed daily, snapshot 2026-11-21.