GitHub

The Quest for Generalizable Motion Generation: Data, Model, and Evaluation

This is the official repository for The Quest for Generalizable Motion Generation: Data, Model, and Evaluation. The repo provides a unified framework for generalizable motion generation, including ...
IEEE

Hard Sample Mining: A New Paradigm of Efficient and Robust Model Training

Abstract: Over the past two decades, deep learning (DL) has achieved unprecedented breakthroughs across diverse application domains spanning computer vision (CV) to natural language processing (NLP).
SlashGear

5 Underappreciated Classic Models From The Early Days Of Volkswagen

The dark history of Volkswagen started on May 28, 1937, when the German government, controlled by Adolf Hitler of the Nazi Party, founded an auto company that was soon renamed as Volkswagenwerk, or ...
Medical Xpress

Researchers discover bias in AI models that analyze pathology samples

Pathology has long been the cornerstone of cancer diagnosis and treatment. A pathologist carefully examines an ultrathin slice of human tissue under a microscope for clues that indicate the presence, ...
Harvard Medical School

Researchers Discover Bias in AI Models That Analyze Pathology Samples

A new study reveals that pathology AI models for cancer diagnosis perform unequally across demographic groups. The researchers identified three explanations for the bias and developed a tool that ...
IEEE

Deep Networks as Denoising Algorithms: Sample-Efficient Learning of Diffusion Models in High-Dimensional Graphical Models

Abstract: We investigate the efficiency of deep neural networks for approximating scoring functions in diffusion-based generative modeling. While existing approximation theories leverage the ...