Researchers baked 3x inference speedups directly into LLM weights — without speculative decoding

Researchers from the University of Maryland, Lawrence Livermore, Columbia and TogetherAI have developed a training technique that triples LLM inference speed without auxiliary models or infrastructure ...
IEEE

A time sequence multimodal deep learning method combining acceleration and frequency domain signal for bearing fault diagnosis

Abstract: The performance of traditional bearing fault diagnosis methods is substantially improved by applying deep learning methods. However, enhancement in the accuracy of prevailing unimodal ...
IEEE

Sequential Deep Learning Model Development for Battery Remaining Useful Life Forecasting and Anomaly Detection

Abstract: Lithium-ion batteries have become indispensable in various energy storage applications, powering a wide array of devices. Anomaly detection and remaining useful life forecasting are critical ...