Mechanical Ventilation Management on Pulse Physiology Engine Severe ARDS cohort
2,240.33Oxygenation Index (r_oxy)VentAgent
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| VentAgentBackbone=gpt-4o-mini2026.06 | 2,240.33 | -42.58 | 175.64 | |
| ReflexionBackbone=gpt-4o-mini2026.06 | 1,995.62 | -102.37 | 132.55 | |
| ToTBackbone=gpt-4o-mini2026.06 | 1,955.15 | -95.82 | 128.46 | |
| DebateBackbone=gpt-4o-mini2026.06 | 1,850.31 | -105.46 | 115.23 | |
| ReActBackbone=gpt-4o-mini2026.06 | 1,745.27 | -110.85 | 108.41 | |
| Self-Consist.Backbone=gpt-4o-mini2026.06 | 1,698.47 | -118.63 | 105.88 | |
| CoTBackbone=gpt-4o-mini2026.06 | 1,605.81 | -128.24 | 95.67 | |
| Few-ShotBackbone=gpt-4o-mini2026.06 | 1,488.45 | -142.56 | 82.11 |