We show that a model owner can artificially introduce uncertainty into their model and provide a corresponding detection mechanism.
-
Updated
Jun 2, 2025 - Jupyter Notebook
We show that a model owner can artificially introduce uncertainty into their model and provide a corresponding detection mechanism.
Deepfake detection with Bayesian uncertainty quantification, selective prediction, and an interactive Streamlit demo.
Trustworthy medical image classification: noise-robust ConvNeXt-Tiny with 83.5% accuracy, calibrated selective prediction, HAM10000 + ISIC 2019.
Code for our paper analyzing the looseness of the upper bound on selective classification performance.
Investigation of how sampling strategies affect Selective Prediction performance in Multi Task Learning
Transform enrichment outputs into verifiable pathway claims via stability distillation, evidence modules, and mechanical PASS/ABSTAIN/FAIL audits.
Reproducible pipeline for silent-failure auditing in ECG accept-sets (MIT-BIH) with Newton–Puiseux onset scoring
Code Repository for SCoRE paper
BoundaryBench: Benchmark + tool-augmented method for boundary containment under GPS noise
Add a description, image, and links to the selective-prediction topic page so that developers can more easily learn about it.
To associate your repository with the selective-prediction topic, visit your repo's landing page and select "manage topics."