Please confirm you are human
This browser or connection looks automated. Press and continuously hold the control for 3 seconds to enable Google-hosted web results and, when separately allowed, AI-assisted answers.
A successful check enables 100 search requests. Interactive access does not authorize scraping, systematic collection, or reuse of search output.
News
DocLang: a markup language for LLMs
4+ hour, 28+ min ago (517+ words) We spoke with Peter Staar, the lead researcher behind IBM’s popular document parser, Docling, about why generative AI needs its own document standard. AI is upending how we live and work, but the data that could really transform how businesses…...
CoFrGeNets replace the ‘bones’ of transformer-based models
1+ mon, 3+ day ago (599+ words) This new approach is a practical and conceptual shift, pointing to lighter-weight generative AI models that perform competitively, and in many cases even better. But what if there were a way to cut down the resources it took to run…...
How training environments can teach AI models to misbehave
1+ mon, 3+ day ago (689+ words) A new study presented at ICML showed that language models trained with reinforcement learning can find and exploit loopholes to maximize reward — at a cost. Imagine you’re a developer who’s been tasked with auditing a large language model for safety…...
Building the future of computing, together
3+ mon, 1+ week ago (357+ words) IBM Research’s partnerships with MIT, ETH Zurich, and the University of Illinois will lead to next-generation AI and quantum algorithms and applications. IBM Research is tackling some of the most challenging and high-risk problems in computing. By working closely with…...
Mid-training is essential for LLM reasoning, IBM study shows
3+ mon, 3+ week ago (581+ words) What happens between pre-training and post-training matters a lot more than people may realize, new IBM research shows. For years, the basic recipe for building a capable large language model was straightforward: train a model on mountains of text, then…...
Meet the IBM researchers trying to make LLMs smarter
6+ mon, 3+ week 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…...