News
Meet the IBM researchers trying to make LLMs smarter
2+ hour, 12+ min ago (853+ words) The tech leads behind IBM's new software library, Mellea, explain how it works and why giving language models explicit requirements can improve performance and lower costs. Large language models still dominate AI leaderboards, but a new class of lighter weight…...
What’s next in computing is generative and quantum
8+ mon, 1+ week ago (457+ words) At Think 2025, IBM Research leaders explored how generative computing swaps AI prompting for programming, and outlined that quantum advantage is just around the corner. Catch up on all of the news and keynotes from this year's Think here. And for…...
IBM crossed a transformer with an SSM and got ‘Bamba’
8+ mon, 3+ week ago (902+ words) In collaboration with CMU, Princeton, and University of Illinois, IBM Research built an open-source LLM that combines the expressive power of a transformer with the runtime speed of a state-space model. Key features will soon be added to IBM Granite…...