Interpretive Canons (BVerfG) - Historical Interpretation: leaderboard

Metric: Positive-class F1 (%; historical interpretation (geschichte): whether the argument rests on the legislative history, sentence-level binary classification in German Federal Constitutional Court decisions; expert-written prompt with worked examples (the same for every model), reasoning models at temperature 1.0; decision-disjoint held-out test split (60% of each subtask, 1:3 positives to negatives)). Source: arxiv.org. Saturation forecast: Around December 2026. 4 models tracked.

Top models

#ModelScore
1Gemini 3.5 Flash Lite72.7
2MiniMax-M364.6
3DeepSeek V4 Pro62.1
4DeepSeek V4 Flash62

Interactive version: theaggregate.ai/benchmark?slug=interpretive-canons-bverfg-historical-interpretation · How It Works · Data refreshed daily, snapshot 2026-09-26.