MedStruct-S (Two-Shot) - Key Discovery: leaderboard
Listing the field keys present on OCR-derived Chinese clinical report pages; MedStruct-S (Li, Wang and Qian, 2026). Micro-averaged F1 in percent on exact key matches, two-shot prompting; higher is better.
Metric: Key F1 (%; exact key match, micro-averaged, OCR-derived Chinese clinical report pages, two-shot prompting, greedy decoding). Source: arxiv.org. Saturation forecast: Estimated already saturated. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 32B Instruct | 56.14 | #1216 |
| 2 | Qwen 3 14B | 48.65 | #1235 |
| 3 | Qwen 3 0.6B | 6.84 | #8392 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=medstruct-s-two-shot-key-discovery · How It Works · Data refreshed daily, snapshot 2026-11-12.