Records found: 4
'Ling 2.0 is a reasoning-first sparse MoE family from Ant Group that keeps per-token compute low with a 1/32 activation recipe while scaling from 16B to 1T parameters.'
Meta AI and Carnegie Mellon researchers unveil UMA, a groundbreaking family of universal atomic models that deliver high accuracy and speed across diverse chemical and materials science tasks without fine-tuning.
Meta and collaborators developed a novel framework to accurately quantify how much language models memorize training data, estimating GPT models store around 3.6 bits per parameter and providing new insights into memorization versus generalization.
Shanghai AI Lab researchers propose entropy-based scaling laws and novel techniques to overcome exploration collapse in reinforcement learning for reasoning-centric large language models, achieving significant performance improvements.