IEEE

Detecting Adversarial Examples Using Cross-Modal Semantic Embeddings From Images and Text

Abstract: Deep learning models are highly susceptible to adversarial attacks, where subtle perturbations in the input images lead to misclassifications. Adversarial examples typically distort specific ...
IEEE

When Pigs Fly: Evaluating Semantic Overconfidence in Deep Neural Network Classifiers

Abstract: We introduce semantic overconfidence as the phenomenon where a model’s output probability remains invariant regardless of the presence or absence of a semantically strong but ...
GitHub

Real-Time Hand Pose Recognition (ASL A–J)

Take a photo of your hand signing an ASL letter (A–J) and the app will detect the hand, draw landmarks, and display the predicted letter. Images are fed through MediaPipe's Hand Landmarker. 21 (x, y) ...
GitHub

Public evaluation data and reproducibility tools for Sentinel alignment seeds.

sentinel-evaluations/ ├── INDEX.json # Central index with all metadata ├── seeds/ # Seed definitions (full text) │ ├── fas-v4/seed.txt # Foundation Alignment Seed v4.1 │ ├── sentinel-v2-thsp/seed.txt ...