Tag: sparse_attention

Media Monitor: New Technologies Emerge to Challenge AI's Transformer Foundation

10 Aug 2026 5:00 a.m. · Source: MIT Technology Review

MIT Technology Review reports that the foundational transformer architecture in large language models is showing its age, prompting startups like Subquadratic, Manifest AI, Liquid AI, Inception, and Pathway to develop new, more efficient AI technologies.

Science & Technology Business Artificial Intelligence Computers and Internet Economy Education Large Language Models Transformers Neural Networks Sparse Attention Diffusion Models State Spaces MIT Technology Review MIT science technology
Media Monitor: AI startup Subquadratic's LLM claims gain traction with independent tests, MIT Technology Review reports

19 Jun 2026 6:40 a.m. · Source: MIT Technology Review

MIT Technology Review reports on Subquadratic, an AI startup claiming a breakthrough in large language models. Independent tests by Appen suggest its SubQ model is faster, cheaper, and more efficient, addressing initial skepticism.

Science & Technology Business Artificial Intelligence Computers and Internet Economy Subquadratic SubQ Large Language Models Sparse Attention Appen MIT Technology Review MIT science technology