LexIssue - Legal Issue Classification Tier-3: leaderboard
Metric: Weighted F1 (%; legal issue classification at Tier-3, the full hierarchical category path, counted only for aligned issues; predicted issues are aligned to the expert reference issues by a two-stage Qwen-3.5-27B judge (semantic retrieval, then rubric-based equivalence verification); mean case-level weighted F1 over the 430 cases; zero-shot prompting with task instructions only). Source: arxiv.org. Saturation forecast: Around April 2028. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash | 35.7 |
| 2 | DeepSeek V4 Flash | 20.4 |
| 3 | GPT-5.2 | 18.5 |
| 4 | Qwen 3 8B | 9.8 |
Interactive version: theaggregate.ai/benchmark?slug=lexissue-legal-issue-classification-tier-3 · How It Works · Data refreshed daily, snapshot 2026-09-26.