Reconstruction (Research Idea Recovery) - Machine Learning: leaderboard
The 120 machine learning papers of Reconstruction (AI-Professor Project, 2026), where a model sees only the pre-publication bibliography of a paper and proposes five hypotheses judged against its withheld idea. Match rate in percent, averaged over leave-one-out peer judges; higher is better.
Metric: Match rate (%; the 120 machine learning seed papers; share of a model's five hypotheses, proposed in one call from a published paper's pre-publication bibliography only (titles and abstracts under anonymous IDs, the paper itself and later literature withheld), that an LLM judge matches to the held-out paper's idea; mean over the six other evaluated models as leave-one-out judges). Source: arxiv.org. Saturation forecast: Around 2029. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.8 | 8.2 | #56 |
| 2 | Kimi K3 | 7.9 | #53 |
| 3 | GPT-5.6 Pro Sol | 7.6 | #25 |
| 4 | GLM-5.2 | 7.2 | #137 |
| 5 | Gemini 3.1 Pro (Preview) | 5.9 | #91 |
| 6 | Qwen 3.7 Max | 4.6 | #142 |
| 7 | DeepSeek V4 Pro | 3.4 | #181 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=reconstruction-research-idea-recovery-machine-learning · How It Works · Data refreshed daily, snapshot 2026-11-12.