ForeSci - Prediction Factuality - Bottleneck-Opportunity Discovery: leaderboard
The 125 Bottleneck-Opportunity Discovery tasks of ForeSci (Tian et al., 2026): from literature published before a cutoff, identify a root bottleneck in a research subdirection and the opportunity reducing it would unlock. Claim-level F1 against hidden validation claims from post-cutoff papers, judged by DeepSeek-V4 and scaled to 0-100 (higher is better), for models answering without retrieval.
Metric: Prediction factuality (claim-level F1 x 100; answer claims supported by, and hidden post-cutoff validation claims covered by, the answer, judged by DeepSeek-V4 with half credit for partial support; 125 tasks naming a root bottleneck and the opportunity it unlocks; native LLM without retrieval, web search disabled). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 235B A22B | 84 | #622 |
| 2 | GPT-5.2 | 79.9 | #172 |
| 3 | GLM-4.6 | 65.1 | #485 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=foresci-prediction-factuality-bottleneck-opportunity-discovery · How It Works · Data refreshed daily, snapshot 2026-11-12.