Area Under The Curve
PulseAugur coverage of Area Under The Curve — every cluster mentioning Area Under The Curve across labs, papers, and developer communities, ranked by signal.
-
AI performance metrics can hide subgroup disparities, study finds
A new research paper published on arXiv explores the concept of collapsibility in performance metrics for clinical predictive AI models. The study identifies that certain metrics, including the area under the receiver o…
-
Face-free dataset of vegetables improves presentation attack detection
Researchers have developed a novel approach to face presentation attack detection (PAD) by training models on a dataset of tomatoes, potatoes, and onions (TPO) instead of faces. This face-free dataset, comprising bona f…
-
AI surveillance benchmarks fail real-world tests, study finds
A new audit of AI surveillance systems reveals that benchmark performance metrics, specifically AUC scores, do not translate to real-world deployability. Researchers found that models trained on one dataset and scene pe…
-
New ALDM model enhances few-shot 3D MRI synthesis for gliomas
Researchers have developed the Anatomically-conditioned Latent Diffusion Model (ALDM), a novel framework designed for efficient, few-shot 3D volumetric MRI synthesis. This model employs a two-stage process, first learni…
-
AI generates synthetic PET scans to improve lung cancer histology classification
Researchers have developed a novel framework using a 3D Pix2Pix Generative Adversarial Network (GAN) to create synthetic PET scans from CT data for non-small cell lung cancer (NSCLC) histology classification. This "virt…
-
Recommender systems should prioritize serendipity over pure accuracy for user engagement.
Accuracy is not the sole metric for evaluating recommender systems, as serendipity—the ability to pleasantly surprise users—is also crucial for long-term engagement. While accuracy metrics like NDCG and MAP are widely a…