NoisyCausal - Causal Swap: leaderboard
Causal reasoning under structured noise (NoisyCausal; Xu and Fu, 2026), causal swap condition: 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, with the values of causally linked variables swapped in the story. Accuracy in percent, higher is better.
Metric: Accuracy (%; synthetic structural-causal-model questions with the values of causally linked variables swapped). Source: arxiv.org. Saturation forecast: Estimated already saturated. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 50.1 | #996 |
| 2 | GPT-3.5 | 44.7 | #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-causal-swap · How It Works · Data refreshed daily, snapshot 2026-11-12.