NoisyCausal - No Noise: leaderboard
Causal reasoning over narrated structural causal models (NoisyCausal; Xu and Fu, 2026) with clean observations: random 3-7 variable causal graphs are grounded in medicine, education or economics scenarios and the model answers an interventional, counterfactual or attributional question whose answer follows from the clean model. Accuracy in percent, higher is better.
Metric: Accuracy (%; synthetic structural-causal-model questions with clean observations). Source: arxiv.org. Saturation forecast: Estimated already saturated. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 62.8 | #996 |
| 2 | GPT-3.5 | 57.9 | #2430 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=noisycausal-no-noise · How It Works · Data refreshed daily, snapshot 2026-11-12.