Beyond Rating: 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 weakness Max-Recall in percent, the share of one human reviewer's weakness points the generated review covers, taking the best-covered reviewer per paper; higher is better.

Metric: Weakness Max-Recall (from 0-1, times 100): share of a single human reviewer's weakness points that the generated review's weakness points cover, taking the best-covered reviewer per paper, on 1,000 test papers from ICLR 2024-2026 and NeurIPS 2022-2025 with three to five high-confidence human reviews each, the model writing a full review from the parsed paper; review points are split into atomic claims and matched to the human reviews' claims by Qwen3-235B; higher is better. Source: arxiv.org. Saturation forecast: Around February 2028. 15 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Claude Sonnet 4.546#261
2GPT-5.242#185
3AgentReview + GPT-540
4AI Scientist + GPT-538
5DeepSeek V3.236#419
6Qwen 3 235B A22B 2507 Instruct32#616
7DeepReviewer-14B31#844
8Qwen 3 30B A3B 2507 Instruct29#1100
9Gemini 3 Pro (Preview)28#119
10SEA-E-7B26
11Qwen 3 8B25#1873
12CycleReviewer-70B20#1880
13OpenReviewer-8B19
14Llama 3.1 8B Instruct16#4600
15Llama 3.1 70B Instruct16#1571

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 · How It Works · Data refreshed daily, snapshot 2026-11-21.