Afin de mieux comprendre ce qui se passe sous le capot, des chercheurs du #MIT et d'ailleurs ont étudié comment les systèmes #iA d'apprentissage du #langage exploitent les données. Et les modèles analysés semblent utiliser un mécanisme étonnamment simple
https://news.mit.edu/2024/large-language-models-use-surprisingly-simple-mechanism-retrieve-stored-knowledge-0325
https://news.mit.edu/2024/large-language-models-use-surprisingly-simple-mechanism-retrieve-stored-knowledge-0325
MIT News
Large language models use a surprisingly simple mechanism to retrieve some stored knowledge
Researchers find large language models use a simple mechanism to retrieve stored knowledge when they respond to a user prompt. These mechanisms can be leveraged to see what the model knows about different subjects and possibly to correct false information…