PRISM-BN - Edge F1: leaderboard
Metric: Edge F1 (%; directed edges whose endpoints were both matched, no credit for reversed edges, so conditional on node alignment; zero-shot extraction of a parameterized Bayesian network from each of the 5,054 PRISM-BN text descriptions through the four-phase prompt pipeline; predicted node and state names are aligned one-to-one with the reference by a Llama 3.3 70B semantic judge). Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V3 | 96.6 |
| 2 | Llama 4 Maverick | 96 |
| 3 | Claude Haiku 4.5 | 95.9 |
| 4 | Qwen 3 30B A3B 2507 Instruct | 93.9 |
| 5 | Gemma 3 12B (IT) | 93.2 |
| 6 | GPT-4o Mini | 90.2 |
Interactive version: theaggregate.ai/benchmark?slug=prism-bn-edge-f1 · How It Works · Data refreshed daily, snapshot 2026-09-26.