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 ...
Morning Overview on MSN

AI cracks 'impossible' math problems, but can it intimidate top geniuses?

Google DeepMind’s AlphaProof system scored at a silver-medal level when tested against the 2024 International Mathematical ...