Patient Status Prediction on RealICU-Scale (test)
51.9AccuracyICU-Evo
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ICU-EvoBackbone=Gemini-3.1-pro [8], System=ICU-Evo2026.05 | 51.9 | 34.8 | |
| RAGBackbone=Gemini-3.1-pro [8], System=RAG2026.05 | 44.2 | 31.2 | |
| ICU-EvoBackbone=GPT-5.4 [22], System=ICU-Evo2026.05 | 43.8 | 32.7 | |
| Local-windowBackbone=GPT-5.4 [22], System=Local-window2026.05 | 41.5 | 26.5 | |
| RAGBackbone=GPT-5.4 [22], System=RAG2026.05 | 41.1 | 26.9 | |
| Local-windowBackbone=Gemini-3.1-pro [8], System=Local-window2026.05 | 40.5 | 26.4 | |
| RAGBackbone=Qwen3-235B [34], System=RAG2026.05 | 36.7 | 28.2 | |
| ICU-EvoBackbone=Qwen3-235B [34], System=ICU-Evo2026.05 | 30.4 | 17.7 | |
| Full-contextBackbone=Qwen3-235B [34], System=Full-context2026.05 | 20.1 | 11.6 | |
| Local-windowBackbone=Qwen3-235B [34], System=Local-window2026.05 | 17.5 | 15.9 |