Future Querying: Can LLMs Serve as Implicit Medical World Models?
2026-08-24 • Computation and Language
Computation and LanguageArtificial Intelligence
AI summaryⓘ
The authors propose a new way to use large language models (LLMs) to predict what might happen to patients in the future by asking time-specific questions based on clinical notes. Their method works directly with unstructured text without needing special features or retraining for different tasks. They show that smaller fine-tuned models can perform close to larger proprietary ones, which is useful for keeping data private. They tested their approach on both made-up medical reports and real ICU notes, finding that LLMs can understand some medical changes over time.
large language modelsclinical predictionunstructured clinical textfine-tuningMIMIC-IVpatient trajectoriesprivacy-preserving deploymenttime-indexed queriessynthetic medical data
Authors
Siri Willems, James Butterworth, Lore Goetschalckx, Peter Vrancx, Philippe Modard, Elke Giets, Ludovic Denoyer
Abstract
Traditional clinical prediction models rely on task-specific pipelines and curated, structured data, which scale poorly and underutilize unstructured text. To address this, we introduce future querying, a paradigm that probes whether large language models (LLMs) can function as implicit medical world models by evaluating their ability to answer time-indexed clinical queries about a patient's future. Our framework operates on unstructured clinical documentation using endpoint-agnostic training, enabling a single model to answer diverse clinical queries over patient trajectories without manual feature engineering or task-specific retraining. We show that small, locally fine-tuned open-weight models can match or approach larger proprietary systems, making the framework suitable for privacy-preserving, on-premise deployment. Evaluated on a new synthetic medical reports dataset and real ICU notes from the MIMIC-IV dataset, our results provide encouraging evidence that LLMs can capture aspects of clinical dynamics.