Clinical Summarization on MIMIC-IV (Quality Metrics)
1.88HallucinationAgenticSum
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| AgenticSum2026.02 | 1.88 | 3.43 | 3.88 | 4.01 | |
| ConTextual2026.02 | 2.08 | 3.42 | 3.82 | 4.55 | |
| Llama-3.2-3B2026.02 | 2.11 | 3.25 | 4.38 | 3.79 | |
| Mistral-7B2026.02 | 2.19 | 3.19 | 4.54 | 3.82 | |
| MedAlpaca-7B2026.02 | 2.21 | 3.18 | 4.8 | 3.96 | |
| Gemma3-1B2026.02 | 2.35 | 3.08 | 4.28 | 3.71 | |
| Flan-T52026.02 | 2.37 | 3.18 | 4.6 | 3.79 | |
| BioBART2026.02 | 2.45 | 2.95 | 4.21 | 3.68 | |
| T5-Large2026.02 | 2.52 | 2.88 | 4.08 | 3.61 |