Mental state language assessment on DementiaBank manual transcripts
2.29Control Mean Score4o-mini
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| 4o-miniModel Version=GPT-4o-mini, Transcript type=Manual2026.06 | 2.29 | 2.5 | 0.09 | 0.31 | 0.25 | 0.23 | 0.19 | |
| 4oModel Version=GPT-4o, Transcript type=Manual2026.06 | 2.24 | 2.59 | 0.15 | 0.29 | 0.29 | 0.39 | 0.29 | |
| LLaMAModel Version=Llama 3 70B, Transcript type=Manual2026.06 | 1.6 | 1.91 | 0.19 | 0.17 | 0.21 | 0.07 | 0.59 | |
| ClaudeModel Version=Claude 3.5 Sonnet, Transcript type=Manual2026.06 | 1.48 | 2.05 | 0.38 | 0.53 | 0.39 | 0.49 | 0.37 |