Retrofitting language models to operate over bytes(nature.com)
99 points by theanonymousone | 20 comments
Open-access Nature research article describing a method to retrofit existing language models to operate directly over bytes instead of tokens.
10 stories tagged nature, newest first.
99 points by theanonymousone | 20 comments
Open-access Nature research article describing a method to retrofit existing language models to operate directly over bytes instead of tokens.
12 points by sbulaev | 0 comments
Nature news article on an epigenetic editing technique that uses chemical tags to silence hepatitis B viral DNA hidden in the host genome.
331 points by DiffTheEnder | 208 comments
Personal blog essay on bird intelligence, prompted by a video of New Zealand kea parrots moving traffic cones near a tunnel by Milford Sound.
401 points by ascorbic | 280 comments
Nature news article about a new antibiotic molecule, found in soil from a technician's garden, that kills drug-resistant bacteria via a novel mechanism.
555 points by kgwgk | 200 comments
Nature news article on a growing movement using AI and large language models to detect errors in research papers, sparked by a flawed black plastic utensils study.
336 points by gmays | 218 comments
Nature news article reporting that after a heart attack, immune cells stimulate sleep neurons in the brain, inducing deep restorative sleep that aids healing.
535 points by rbanffy | 302 comments
A Nature technology feature on why researchers are running small, openly available AI models locally on laptops, with guidance on getting started.
612 points by sohkamyung | 92 comments
Nature news article on Japan allocating 10 billion yen to institutional repositories to make the nation's publicly funded research free to read.
424 points by sohkamyung | 303 comments
Nature news feature on how the global campaign to eradicate polio may soon succeed and the challenge of keeping the disease from returning.
721 points by anjneymidha | 250 comments
Open-access Nature paper from DeepMind describing AlphaDev, which used deep reinforcement learning to discover faster sorting algorithms.