Uncertainty based selection of compatible inputs

Behavioral Trust Clustering a thermodynamic governance layer that reduces LLM hallucination by 52% on HumanEval. Drop-in wrapper for any decoder. MIT.

We show that a model owner can artificially introduce uncertainty into their model and provide a corresponding detection mechanism.

BoundaryBench: Benchmark + tool-augmented method for boundary containment under GPS noise

Reproducible MEDAI deferral simulation (AIRI 2026). Synthetic research code.

Investigation of how sampling strategies affect Selective Prediction performance in Multi Task Learning

Code for our paper analyzing the looseness of the upper bound on selective classification performance.

Deepfake detection with Bayesian uncertainty quantification, selective prediction, and an interactive Streamlit demo.

Code Repository for SCoRE paper

A comprehensive library for uncertainty quantification in machine learning.

Reproducible pipeline for silent-failure auditing in ECG accept-sets (MIT-BIH) with Newton–Puiseux onset scoring

Transform enrichment outputs into verifiable pathway claims via stability distillation, evidence modules, and mechanical PASS/ABSTAIN/FAIL audits.

Trustworthy medical image classification: noise-robust ConvNeXt-Tiny with 83.5% accuracy, calibrated selective prediction, HAM10000 + ISIC 2019.