Warning: mkdir(): No space left on device in /var/www/group-telegram/post.php on line 37
Warning: file_put_contents(aCache/aDaily/post/datastorieslanguages/--): Failed to open stream: No such file or directory in /var/www/group-telegram/post.php on line 50 Data, Stories and Languages | Telegram Webview: datastorieslanguages/351 -
Training Large Language Models to Reason in a Continuous Latent Space
Новая статья от META - про кокосик! То есть Coconut (Chain of Continuous Thought).
Авторы предлагают изменить подход к reasoning в LLM, перемещая процесс из "language space" в "latent space". По сути, модель думает не токенами, а с использованием hidden state. Это позволяет делать breadth-first search и избегать преждевременных решений при выборе неоптимального пути. Coconut превосходит CoT в задачах логического мышления с необходимостью сложного планирования и backtracking.
Подобные идеи уже пробовали в других работах, но у META получилось довольно красиво. Кстати, в качестве базовой модели используют старый добрый GPT-2.
Training Large Language Models to Reason in a Continuous Latent Space
Новая статья от META - про кокосик! То есть Coconut (Chain of Continuous Thought).
Авторы предлагают изменить подход к reasoning в LLM, перемещая процесс из "language space" в "latent space". По сути, модель думает не токенами, а с использованием hidden state. Это позволяет делать breadth-first search и избегать преждевременных решений при выборе неоптимального пути. Coconut превосходит CoT в задачах логического мышления с необходимостью сложного планирования и backtracking.
Подобные идеи уже пробовали в других работах, но у META получилось довольно красиво. Кстати, в качестве базовой модели используют старый добрый GPT-2.
"We're seeing really dramatic moves, and it's all really tied to Ukraine right now, and in a secondary way, in terms of interest rates," Octavio Marenzi, CEO of Opimas, told Yahoo Finance Live on Thursday. "This war in Ukraine is going to give the Fed the ammunition, the cover that it needs, to not raise interest rates too quickly. And I think Jay Powell is a very tepid sort of inflation fighter and he's not going to do as much as he needs to do to get that under control. And this seems like an excuse to kick the can further down the road still and not do too much too soon." During the operations, Sebi officials seized various records and documents, including 34 mobile phones, six laptops, four desktops, four tablets, two hard drive disks and one pen drive from the custody of these persons. These administrators had built substantial positions in these scrips prior to the circulation of recommendations and offloaded their positions subsequent to rise in price of these scrips, making significant profits at the expense of unsuspecting investors, Sebi noted. The Russian invasion of Ukraine has been a driving force in markets for the past few weeks. Although some channels have been removed, the curation process is considered opaque and insufficient by analysts.
from nl
