auprc-4d9aa7c3·3 events·first seen Aliases: AUPRC
A new arXiv preprint develops a theoretical framework characterizing when synthetic minority-class augmentation improves score-based metrics (AUROC, AUPRC, balanced accuracy, F1) under class imbalance. The authors show that under well-specified score models, augmentation provides no fundamental population-level improvement and may introduce bias, while under model misspecification it can correct ranking errors by shifting effective class balance. Minimax lower bounds confirm the raw estimator is already optimal in the well-specified regime, and simulation studies corroborate the theory.
Researchers developed a three-stage pipeline combining traditional machine learning with LLM-based screening and evidence extraction to detect self-harm in Australian emergency department triage notes. The system achieved AUPRCs around 0.88 in both internal and external validation, and transferred to two external hospital sites without site-specific retraining. A notable capability is identifying the primary self-harm method with 95% accuracy, enabling more granular public health surveillance beyond binary classification.
The paper introduces Reverse Probing, a novel uncertainty quantification framework designed specifically for clinical text summarization that estimates token-level uncertainty from pre-existing labeled summaries rather than sampling new outputs. It extracts uncertainty signals from four categories of internal model activations, treating text as a probe into the model's internal state. Evaluated on two expert-annotated clinical datasets, it outperforms eight adapted baselines on all metrics, achieving up to 4× higher AUPRC while reducing inference time and compute. Feature analysis identifies delta energy and neighborhood context as the most consistent predictors of uncertainty across models.