Risk evaluation in caregiving responses on General-Caregiver Dataset
0.08RubRIXGPT-4o-mini
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| GPT-4o-miniRefinement Stage=Turn 22026.01 | 0.08 | -20 | 0.51 | -0.02 | |
| ClaudeRefinement Stage=Turn 22026.01 | 0.09 | -50 | 1.3 | -0.07 | |
| GPT-4o-miniRefinement Stage=Turn 12026.01 | 0.1 | -98 | 46.5 | -2.1 | |
| ClaudeRefinement Stage=Turn 12026.01 | 0.17 | -95 | 33.63 | -1.83 | |
| Phi-3Refinement Stage=Turn 22026.01 | 0.34 | -25 | 3.44 | -0.04 | |
| Phi-3Refinement Stage=Turn 12026.01 | 0.41 | -96.67 | 176.01 | -1.89 | |
| QwenRefinement Stage=Turn 22026.01 | 3 | -25 | 1.22 | -0.04 | |
| MedichatRefinement Stage=Turn 22026.01 | 3 | -40 | 5.51 | -0.12 | |
| ClaudeRefinement Stage=Initial2026.01 | 4 | — | — | — | |
| QwenRefinement Stage=Turn 12026.01 | 4 | -20 | 7.33 | -0.25 | |
| MedalpacaRefinement Stage=Turn 22026.01 | 4 | -20 | 3.03 | -0.05 | |
| GPT-4o-miniRefinement Stage=Initial2026.01 | 5 | — | — | — | |
| QwenRefinement Stage=Initial2026.01 | 5 | — | — | — | |
| MedichatRefinement Stage=Turn 12026.01 | 5 | -16.67 | 6.26 | -0.14 | |
| MedalpacaRefinement Stage=Turn 12026.01 | 5 | -50 | 31.89 | -0.53 | |
| MedichatRefinement Stage=Initial2026.01 | 6 | — | — | — | |
| MedalpacaRefinement Stage=Initial2026.01 | 10 | — | — | — | |
| Phi-3Refinement Stage=Initial2026.01 | 12 | — | — | — |