EPPCMinerBen (0-Shot) - Code Classification: leaderboard
Metric: Micro-averaged F1 (%) of EPPC code classification (sentence-level multi-label assignment of the nine EPPC communication codes, predicted label sets compared with the expert labels), zero-shot (task description only), mean over seeded runs, on 1,933 expert-annotated sentences from 752 Yale New Haven Health patient-portal secure messages; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 2 27B (IT) | 64.62 | #633 |
| 2 | Llama 3.1 70B Instruct | 64.57 | #548 |
| 3 | Llama 3.3 70B Instruct | 64.21 | #520 |
| 4 | DeepSeek R1 Distill Qwen 32B | 61.52 | #640 |
| 5 | Qwen2.5-14B-Instruct-1M | 60.63 | #566 |
| 6 | DeepSeek R1 Distill Llama 70B | 59.43 | #582 |
| 7 | Llama 3.2 3B Instruct | 30.3 | #1321 |
| 8 | Llama 3.2 1B Instruct | 27.16 | #1518 |
| 9 | Qwen 2.5 1.5B Instruct | 25.88 | #1388 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=eppcminerben-0-shot-code-classification · How It Works · Data refreshed daily, snapshot 2026-10-11.