Multi-label Phenotyping on Dementia Assessment
74.3AUCMacroDementia-Agents
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Dementia-AgentsArchitecture=Multi-Agent2026.06 | 74.3 | 24.2 | 36.4 | 75.4 | |
| LLaVA-Med-7BEvaluation Protocol=Supervised Fine-Tuning, Architecture=Single MLLM2026.06 | 63.7 | 16.1 | 25.4 | 60.9 | |
| InternVL3.5-8BEvaluation Protocol=Supervised Fine-Tuning, Architecture=Single MLLM2026.06 | 63.1 | 15.6 | 26 | 63.8 | |
| Hulu-Med-7BEvaluation Protocol=Supervised Fine-Tuning, Architecture=Single MLLM2026.06 | 61.6 | 16.3 | 25.5 | 66.7 | |
| Qwen3-VL-8BEvaluation Protocol=Supervised Fine-Tuning, Architecture=Single MLLM2026.06 | 61.5 | 14.9 | 24.9 | 66.7 | |
| LLaVA-Med-7BEvaluation Protocol=Zero-Shot, Architecture=Single MLLM2026.06 | 57.6 | 16.8 | 18.6 | 43.5 | |
| MedAgentProArchitecture=Multi-Agent2026.06 | 52.5 | 12.1 | 30.8 | 55.1 | |
| InternVL3.5-8BEvaluation Protocol=Zero-Shot, Architecture=Single MLLM2026.06 | 51 | 6.8 | 22.2 | 43.5 | |
| MedAgentsArchitecture=Multi-Agent2026.06 | 50.8 | 19 | 21.9 | 59.4 | |
| Dementia-AgentsArchitecture=Multi-Agent, Evaluation Protocol=Zero-Shot2026.06 | 50.5 | 17.9 | 21 | 36.2 | |
| Qwen3-VL-8BEvaluation Protocol=Zero-Shot, Architecture=Single MLLM2026.06 | 50.1 | 10.9 | 14.5 | 31.9 | |
| Hulu-Med-7BEvaluation Protocol=Zero-Shot, Architecture=Single MLLM2026.06 | 46.8 | 13.4 | 14.9 | 17.4 |