CausaLab - Graph SHD (6-Node SCMs): leaderboard
Metric: Directed structural Hamming distance of the final hypothesised graph to the hidden 6-node SCM (missing, extra and reversed edges; mean per episode, lower is better). Source: arxiv.org. Saturation forecast: Around October 2027. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 (High) | 2.66 |
| 2 | GPT-5 Mini | 5.49 |
Interactive version: theaggregate.ai/benchmark?slug=causalab-graph-shd-6-node-scms · How It Works · Data refreshed daily, snapshot 2026-09-26.