RealICU (RAG) - Acute Problems: leaderboard
RealICU (Shen et al., 2026) replays MIMIC-IV intensive-care stays in 30-minute windows and scores an LLM against hindsight labels from senior ICU physicians on the 930 windows of RealICU-Gold. Acute Problems: Hit@5 (0-1), the share of windows whose top five predicted acute problems include a physician-labelled one under PubMedBERT semantic matching, higher is better. RAG protocol: the model sees the current window and the five most relevant windows of the stay so far, retrieved with PubMedBERT embeddings.
Metric: Hit@5 (0-1; share of windows whose top-5 predicted acute problems include a physician label, PubMedBERT semantic matching; RAG strategy, the top-5 prior windows retrieved with PubMedBERT). Source: arxiv.org. Saturation forecast: Around February 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 0.6 | #121 |
| 2 | Gemini 3.1 Pro (Preview) | 0.6 | #91 |
| 3 | Qwen 3 235B A22B | 0.38 | #622 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=realicu-rag-acute-problems · How It Works · Data refreshed daily, snapshot 2026-11-12.