Reconstruction (Research Idea Recovery): leaderboard

Metric: Match rate (%; unweighted mean of the six domain scores over 643 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

#ModelScore
1Claude Opus 4.813.3
2GPT-5.6 Pro Sol12.8
3Kimi K310
4GLM-5.29.3
5Gemini 3.1 Pro (Preview)8.9
6DeepSeek V4 Pro6.3
7Qwen 3.7 Max5.9

Interactive version: theaggregate.ai/benchmark?slug=reconstruction-research-idea-recovery · How It Works · Data refreshed daily, snapshot 2026-09-26.