RINoBench - Justification Alignment: leaderboard
Metric: Justification alignment (ALI, 0-1 G-Eval score from an LLM judge, shown times 100): whether the model's novelty justification follows the same arguments and conclusion as the gold justification synthesized from the human reviews, on the RINoBench test split (one fifth of the 1,381 ICLR 2022-2023 research ideas, gold scores from human reviewers), zero-shot with the five-point novelty rubric and the idea's related works; higher is better. Source: arxiv.org. Saturation forecast: Around June 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 | 72 | #121 |
| 2 | GPT-5 | 71 | #91 |
| 3 | DeepSeek R1 | 67 | #245 |
| 4 | GPT-OSS-120B | 64 | #330 |
| 5 | Llama 3.1 8B | 58 | #1139 |
| 6 | Llama 4 Scout | 58 | #646 |
| 7 | Llama 3.3 70B Instruct | 55 | #520 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=rinobench-justification-alignment · How It Works · Data refreshed daily, snapshot 2026-10-11.