Lamine Diop

SEP 3, 2026 · PREPRINT

Differentiable Interval Bottlenecks for Interpretable Anomaly Detection in Numerical Data

arXiv

A novel machine-learning method validated on 48 benchmarks with competitive performance, but presented as a preprint without peer review, making empirical claims about anomaly detection rather than clinical or regulatory evidence.

SEP 3, 2026 · PREPRINT

Witnesses Explain Anomalies

arXiv

A novel computational method for unsupervised anomaly detection with integrated explainability, demonstrated on 47 benchmark datasets but lacking clinical validation, real-world deployment evidence, or peer review.