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

Efficient Hardware Implementation of Partial Response Equalization with Reduced-State Sequence Estimation Decoder and Blind Adaptation Using Delay Zero Forcing Algorithm for DCI

Abstract: With the growing demand for data center interconnect (DCI) capacity, partial response equalization (PRE), employing either a 1/(1+D) decoder or an maximum likelihood sequence estimation ...
The New York Times

Researchers Are Using A.I. to Decode the Human Genome

AlphaGenome is a leap forward in the ability to study the human blueprint. But the fine workings of our DNA are still largely a mystery. By Carl Zimmer In 2024, two scientists from Google DeepMind ...