TTT-Discover optimizes GPU kernels 2x faster than human experts — by training during inference

A new technique from Stanford, Nvidia, and Together AI lets models learn during inference rather than relying on static ...
IEEE

Iterative Learning Control Design for Iteration-Varying State-of-Charge Profiles of Electric Vehicle Batteries

Abstract: In this research, a battery control method is proposed to handle the repetitive but nonidentical daily state-of-charge (SoC) profiles of electric vehicle (EV) batteries. The proposed method ...
IEEE

From Vineyard Data to Flavour Profiles: Machine Learning Predicts Wine Quality

Abstract: This paper looks at how well three machine learning models can predict wine quality. The models being compared are K Nearest Neighbors, Gradient Boosting, and Extreme Gradient Boosting. They ...
GitHub

yyl-github-1896/CodeTAE

This is the official code repository for the paper "Exploiting the Adversarial Example Vulnerability of Transfer Learning of Source Code"(TIFS 2024). State-of-the-art source code classification models ...
Frontiers

Open learner models and pedagogical strategies in higher education: a meta-synthesis of approaches to self-regulated learning

Open Learner Models (OLMs) are increasingly positioned as pedagogical tools capable of enhancing self-regulated learning (SRL) within higher education. Despite considerable technological progress, the ...