11 papers
Extending Pretrained 10-Second ECG Foundation Models to Longer Horizons
Wei Tang, Jinpei Han, Kangning Cui +10
Electrocardiogram (ECG) foundation models pretrained on typical diagnostic 10-second ECG segments, have demonstrated strong transferability across a range of clinical applications.…
BAS: A Decision-Theoretic Approach to Evaluating Large Language Model Confidence
Sean Wu, Fredrik K. Gustafsson, Edward Phillips +3
Large language models (LLMs) often produce confident but incorrect answers in settings where abstention would be safer. Standard evaluation protocols, however, require a response a…
Entropy Alone is Insufficient for Safe Selective Prediction in LLMs
Edward Phillips, Fredrik K. Gustafsson, Sean Wu +2
Selective prediction systems can mitigate harms resulting from language model hallucinations by abstaining from answering in high-risk cases. Uncertainty quantification techniques…
Democratising Clinical AI through Dataset Condensation for Classical Clinical Models
Anshul Thakur, Soheila Molaei, Pafue Christy Nganjimi +5
Dataset condensation (DC) learns a compact synthetic dataset that enables models to match the performance of full-data training, prioritising utility over distributional fidelity.…
RiskAgent: Synergizing Language Models with Validated Tools for Evidence-Based Risk Prediction
Fenglin Liu, Jinge Wu, Hongjian Zhou +9
Large Language Models (LLMs) achieve competitive results compared to human experts in medical examinations. However, it remains a challenge to apply LLMs to complex clinical decisi…
Geometric Uncertainty for Detecting and Correcting Hallucinations in LLMs
Edward Phillips, Sean Wu, Soheila Molaei +3
Large language models demonstrate impressive results across diverse tasks but are still known to hallucinate, generating linguistically plausible but incorrect answers to questions…