{"id":11172,"date":"2024-04-02T09:32:07","date_gmt":"2024-04-02T09:32:07","guid":{"rendered":"https:\/\/dailyai.com\/?p=11172"},"modified":"2024-04-02T09:32:07","modified_gmt":"2024-04-02T09:32:07","slug":"deepmind-developed-safe-an-ai-agent-to-fact-check-llms","status":"publish","type":"post","link":"https:\/\/dailyai.com\/ru\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/","title":{"rendered":"DeepMind \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u0430\u043b\u0430 SAFE, \u0418\u0418-\u0430\u0433\u0435\u043d\u0442 \u0434\u043b\u044f \u043f\u0440\u043e\u0432\u0435\u0440\u043a\u0438 \u0444\u0430\u043a\u0442\u043e\u0432 \u0432 LLM"},"content":{"rendered":"

\u0418\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u0442\u0435\u043b\u0438 \u0438\u0437 DeepMind \u0438 \u0421\u0442\u044d\u043d\u0444\u043e\u0440\u0434\u0441\u043a\u043e\u0433\u043e \u0443\u043d\u0438\u0432\u0435\u0440\u0441\u0438\u0442\u0435\u0442\u0430 \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u0430\u043b\u0438 \u0418\u0418-\u0430\u0433\u0435\u043d\u0442, \u043a\u043e\u0442\u043e\u0440\u044b\u0439 \u043f\u0440\u043e\u0432\u0435\u0440\u044f\u0435\u0442 \u0444\u0430\u043a\u0442\u043e\u043b\u043e\u0433\u0438\u044e LLM \u0438 \u043f\u043e\u0437\u0432\u043e\u043b\u044f\u0435\u0442 \u043f\u0440\u043e\u0432\u043e\u0434\u0438\u0442\u044c \u0441\u0440\u0430\u0432\u043d\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u0439 \u0430\u043d\u0430\u043b\u0438\u0437 \u0444\u0430\u043a\u0442\u043e\u043b\u043e\u0433\u0438\u0447\u043d\u043e\u0441\u0442\u0438 \u0418\u0418-\u043c\u043e\u0434\u0435\u043b\u0435\u0439.<\/strong><\/p>\n

\u0414\u0430\u0436\u0435 \u0441\u0430\u043c\u044b\u0435 \u043b\u0443\u0447\u0448\u0438\u0435 \u043c\u043e\u0434\u0435\u043b\u0438 \u0438\u0441\u043a\u0443\u0441\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0433\u043e \u0438\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0430 \u043f\u043e-\u043f\u0440\u0435\u0436\u043d\u0435\u043c\u0443 \u0441\u043a\u043b\u043e\u043d\u043d\u044b \u0433\u0430\u043b\u043b\u044e\u0446\u0438\u043d\u0438\u0440\u043e\u0432\u0430\u0442\u044c<\/a> \u0438\u043d\u043e\u0433\u0434\u0430. \u0415\u0441\u043b\u0438 \u0432\u044b \u043f\u043e\u043f\u0440\u043e\u0441\u0438\u0442\u0435 ChatGPT \u043f\u0440\u0435\u0434\u043e\u0441\u0442\u0430\u0432\u0438\u0442\u044c \u0432\u0430\u043c \u0444\u0430\u043a\u0442\u044b \u043f\u043e \u0442\u0435\u043c\u0435, \u0442\u043e \u0447\u0435\u043c \u0434\u043b\u0438\u043d\u043d\u0435\u0435 \u0431\u0443\u0434\u0435\u0442 \u043e\u0442\u0432\u0435\u0442, \u0442\u0435\u043c \u0431\u043e\u043b\u044c\u0448\u0435 \u0432\u0435\u0440\u043e\u044f\u0442\u043d\u043e\u0441\u0442\u044c \u0442\u043e\u0433\u043e, \u0447\u0442\u043e \u0432 \u043d\u0435\u043c \u043e\u043a\u0430\u0436\u0443\u0442\u0441\u044f \u043d\u0435\u043f\u0440\u0430\u0432\u0434\u0438\u0432\u044b\u0435 \u0444\u0430\u043a\u0442\u044b.<\/p>\n

\u041a\u0430\u043a\u0438\u0435 \u043c\u043e\u0434\u0435\u043b\u0438 \u0431\u043e\u043b\u0435\u0435 \u0442\u043e\u0447\u043d\u044b \u0432 \u0444\u0430\u043a\u0442\u0430\u0445, \u0447\u0435\u043c \u0434\u0440\u0443\u0433\u0438\u0435, \u043f\u0440\u0438 \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u0438 \u0434\u043b\u0438\u043d\u043d\u044b\u0445 \u043e\u0442\u0432\u0435\u0442\u043e\u0432? \u0421\u043b\u043e\u0436\u043d\u043e \u0441\u043a\u0430\u0437\u0430\u0442\u044c, \u043f\u043e\u0442\u043e\u043c\u0443 \u0447\u0442\u043e \u0434\u043e \u0441\u0438\u0445 \u043f\u043e\u0440 \u0443 \u043d\u0430\u0441 \u043d\u0435 \u0431\u044b\u043b\u043e \u044d\u0442\u0430\u043b\u043e\u043d\u0430, \u0438\u0437\u043c\u0435\u0440\u044f\u044e\u0449\u0435\u0433\u043e \u0444\u0430\u043a\u0442\u0438\u0447\u043d\u043e\u0441\u0442\u044c \u0434\u043b\u0438\u043d\u043d\u044b\u0445 \u043e\u0442\u0432\u0435\u0442\u043e\u0432 LLM.<\/p>\n

\u0421\u043d\u0430\u0447\u0430\u043b\u0430 DeepMind \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043b\u0430 GPT-4 \u0434\u043b\u044f \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u044f LongFact - \u043d\u0430\u0431\u043e\u0440\u0430 \u0438\u0437 2280 \u043f\u043e\u0434\u0441\u043a\u0430\u0437\u043e\u043a \u0432 \u0432\u0438\u0434\u0435 \u0432\u043e\u043f\u0440\u043e\u0441\u043e\u0432, \u043e\u0442\u043d\u043e\u0441\u044f\u0449\u0438\u0445\u0441\u044f \u043a 38 \u0442\u0435\u043c\u0430\u043c. \u042d\u0442\u0438 \u043f\u043e\u0434\u0441\u043a\u0430\u0437\u043a\u0438 \u0432\u044b\u0437\u044b\u0432\u0430\u044e\u0442 \u0443 \u0442\u0435\u0441\u0442\u0438\u0440\u0443\u0435\u043c\u043e\u0433\u043e LLM \u0440\u0430\u0437\u0432\u0435\u0440\u043d\u0443\u0442\u044b\u0435 \u043e\u0442\u0432\u0435\u0442\u044b.<\/p>\n

\u0417\u0430\u0442\u0435\u043c \u043e\u043d\u0438 \u0441\u043e\u0437\u0434\u0430\u043b\u0438 \u0418\u0418-\u0430\u0433\u0435\u043d\u0442, \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u044e\u0449\u0438\u0439 GPT-3.5-turbo, \u0447\u0442\u043e\u0431\u044b \u0441 \u043f\u043e\u043c\u043e\u0449\u044c\u044e Google \u043f\u0440\u043e\u0432\u0435\u0440\u0438\u0442\u044c, \u043d\u0430\u0441\u043a\u043e\u043b\u044c\u043a\u043e \u0444\u0430\u043a\u0442\u0438\u0447\u043d\u044b \u043e\u0442\u0432\u0435\u0442\u044b, \u0441\u0433\u0435\u043d\u0435\u0440\u0438\u0440\u043e\u0432\u0430\u043d\u043d\u044b\u0435 LLM. \u041e\u043d\u0438 \u043d\u0430\u0437\u0432\u0430\u043b\u0438 \u044d\u0442\u043e\u0442 \u043c\u0435\u0442\u043e\u0434 Search-Augmented Factuality Evaluator (SAFE).<\/p>\n

SAFE \u0441\u043d\u0430\u0447\u0430\u043b\u0430 \u0440\u0430\u0437\u0431\u0438\u0432\u0430\u0435\u0442 \u0434\u043b\u0438\u043d\u043d\u044b\u0439 \u043e\u0442\u0432\u0435\u0442 LLM \u043d\u0430 \u043e\u0442\u0434\u0435\u043b\u044c\u043d\u044b\u0435 \u0444\u0430\u043a\u0442\u044b. \u0417\u0430\u0442\u0435\u043c \u043e\u043d \u043e\u0442\u043f\u0440\u0430\u0432\u043b\u044f\u0435\u0442 \u043f\u043e\u0438\u0441\u043a\u043e\u0432\u044b\u0435 \u0437\u0430\u043f\u0440\u043e\u0441\u044b \u0432 Google Search \u0438 \u043e\u043f\u0440\u0435\u0434\u0435\u043b\u044f\u0435\u0442 \u043f\u0440\u0430\u0432\u0434\u0438\u0432\u043e\u0441\u0442\u044c \u0444\u0430\u043a\u0442\u0430 \u043d\u0430 \u043e\u0441\u043d\u043e\u0432\u0435 \u0438\u043d\u0444\u043e\u0440\u043c\u0430\u0446\u0438\u0438, \u043f\u043e\u043b\u0443\u0447\u0435\u043d\u043d\u043e\u0439 \u0432 \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\u0430\u0445 \u043f\u043e\u0438\u0441\u043a\u0430.<\/p>\n

\u0412\u043e\u0442 \u043f\u0440\u0438\u043c\u0435\u0440 \u0438\u0437 \u043d\u0430\u0443\u0447\u043d\u0430\u044f \u0441\u0442\u0430\u0442\u044c\u044f<\/a>.<\/p>\n

\"\"
\u0417\u0430\u0434\u0430\u043d\u0438\u0435 \u043d\u0430 \u043f\u043e\u0438\u0441\u043a \u0444\u0430\u043a\u0442\u043e\u0432 \u043f\u043e\u0437\u0432\u043e\u043b\u044f\u0435\u0442 \u043f\u043e\u043b\u0443\u0447\u0438\u0442\u044c \u0440\u0430\u0437\u0432\u0435\u0440\u043d\u0443\u0442\u044b\u0439 \u043e\u0442\u0432\u0435\u0442. \u041e\u0442\u0432\u0435\u0442 \u0440\u0430\u0437\u0431\u0438\u0432\u0430\u0435\u0442\u0441\u044f \u043d\u0430 \u043e\u0442\u0434\u0435\u043b\u044c\u043d\u044b\u0435 \u0444\u0430\u043a\u0442\u044b, \u043f\u0435\u0440\u0435\u0441\u043c\u0430\u0442\u0440\u0438\u0432\u0430\u0435\u0442\u0441\u044f, \u0447\u0442\u043e\u0431\u044b \u0431\u044b\u0442\u044c \u0441\u0430\u043c\u043e\u0441\u0442\u043e\u044f\u0442\u0435\u043b\u044c\u043d\u044b\u043c, \u043f\u0440\u043e\u0432\u0435\u0440\u044f\u0435\u0442\u0441\u044f \u043d\u0430 \u0440\u0435\u043b\u0435\u0432\u0430\u043d\u0442\u043d\u043e\u0441\u0442\u044c \u0438 \u043f\u0440\u043e\u0432\u0435\u0440\u044f\u0435\u0442\u0441\u044f \u0441 \u043f\u043e\u043c\u043e\u0449\u044c\u044e Google Search. \u0418\u0441\u0442\u043e\u0447\u043d\u0438\u043a: arXiv<\/figcaption><\/figure>\n

\u0418\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u0442\u0435\u043b\u0438 \u0443\u0442\u0432\u0435\u0440\u0436\u0434\u0430\u044e\u0442, \u0447\u0442\u043e SAFE \u0434\u043e\u0441\u0442\u0438\u0433\u0430\u0435\u0442 \"\u0441\u0432\u0435\u0440\u0445\u0447\u0435\u043b\u043e\u0432\u0435\u0447\u0435\u0441\u043a\u043e\u0439 \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u0438\" \u043f\u043e \u0441\u0440\u0430\u0432\u043d\u0435\u043d\u0438\u044e \u0441 \u0447\u0435\u043b\u043e\u0432\u0435\u0447\u0435\u0441\u043a\u0438\u043c\u0438 \u0430\u043d\u043d\u043e\u0442\u0430\u0442\u043e\u0440\u0430\u043c\u0438, \u0432\u044b\u043f\u043e\u043b\u043d\u044f\u044e\u0449\u0438\u043c\u0438 \u043f\u0440\u043e\u0432\u0435\u0440\u043a\u0443 \u0444\u0430\u043a\u0442\u043e\u0432.<\/p>\n

SAFE \u0441\u043e\u0433\u043b\u0430\u0441\u0438\u043b\u0441\u044f \u0441 72% \u0447\u0435\u043b\u043e\u0432\u0435\u0447\u0435\u0441\u043a\u0438\u0445 \u0430\u043d\u043d\u043e\u0442\u0430\u0446\u0438\u0439, \u0430 \u0442\u0430\u043c, \u0433\u0434\u0435 \u0435\u0433\u043e \u043c\u043d\u0435\u043d\u0438\u0435 \u0440\u0430\u0441\u0445\u043e\u0434\u0438\u043b\u043e\u0441\u044c \u0441 \u0447\u0435\u043b\u043e\u0432\u0435\u0447\u0435\u0441\u043a\u0438\u043c, \u043e\u043d \u043e\u043a\u0430\u0437\u0430\u043b\u0441\u044f \u043f\u0440\u0430\u0432 \u0432 76% \u0441\u043b\u0443\u0447\u0430\u0435\u0432. \u041a\u0440\u043e\u043c\u0435 \u0442\u043e\u0433\u043e, \u044d\u0442\u0430 \u0441\u0438\u0441\u0442\u0435\u043c\u0430 \u043e\u043a\u0430\u0437\u0430\u043b\u0430\u0441\u044c \u0432 20 \u0440\u0430\u0437 \u0434\u0435\u0448\u0435\u0432\u043b\u0435, \u0447\u0435\u043c \u043a\u0440\u0430\u0443\u0434\u0441\u043e\u0440\u0441\u0438\u043d\u0433\u043e\u0432\u044b\u0435 \u0447\u0435\u043b\u043e\u0432\u0435\u0447\u0435\u0441\u043a\u0438\u0435 \u0430\u043d\u043d\u043e\u0442\u0430\u0442\u043e\u0440\u044b. \u0422\u0430\u043a\u0438\u043c \u043e\u0431\u0440\u0430\u0437\u043e\u043c, LLM \u043b\u0443\u0447\u0448\u0435 \u0438 \u0434\u0435\u0448\u0435\u0432\u043b\u0435 \u043f\u0440\u043e\u0432\u0435\u0440\u044f\u044e\u0442 \u0444\u0430\u043a\u0442\u044b, \u0447\u0435\u043c \u043b\u044e\u0434\u0438.<\/p>\n

\u041a\u0430\u0447\u0435\u0441\u0442\u0432\u043e \u043e\u0442\u0432\u0435\u0442\u0430 \u0438\u0441\u043f\u044b\u0442\u0443\u0435\u043c\u044b\u0445 LLM \u043e\u0446\u0435\u043d\u0438\u0432\u0430\u043b\u043e\u0441\u044c \u043f\u043e \u043a\u043e\u043b\u0438\u0447\u0435\u0441\u0442\u0432\u0443 \u0444\u0430\u043a\u0442\u043e\u0438\u0434\u043e\u0432 \u0432 \u043e\u0442\u0432\u0435\u0442\u0435, \u0430 \u0442\u0430\u043a\u0436\u0435 \u043f\u043e \u0442\u043e\u043c\u0443, \u043d\u0430\u0441\u043a\u043e\u043b\u044c\u043a\u043e \u0444\u0430\u043a\u0442\u043e\u0438\u0434\u0430\u043c\u0438 \u0431\u044b\u043b\u0438 \u043e\u0442\u0434\u0435\u043b\u044c\u043d\u044b\u0435 \u0444\u0430\u043a\u0442\u044b.<\/p>\n

\u0418\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u0435\u043c\u0430\u044f \u0438\u043c\u0438 \u043c\u0435\u0442\u0440\u0438\u043a\u0430 (F1@K) \u043e\u0446\u0435\u043d\u0438\u0432\u0430\u0435\u0442 \u043f\u0440\u0435\u0434\u043f\u043e\u0447\u0442\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0435 \u0434\u043b\u044f \u0447\u0435\u043b\u043e\u0432\u0435\u043a\u0430 \"\u0438\u0434\u0435\u0430\u043b\u044c\u043d\u043e\u0435\" \u043a\u043e\u043b\u0438\u0447\u0435\u0441\u0442\u0432\u043e \u0444\u0430\u043a\u0442\u043e\u0432 \u0432 \u043e\u0442\u0432\u0435\u0442\u0435. \u0412 \u044d\u0442\u0430\u043b\u043e\u043d\u043d\u044b\u0445 \u0442\u0435\u0441\u0442\u0430\u0445 \u0432 \u043a\u0430\u0447\u0435\u0441\u0442\u0432\u0435 \u043c\u0435\u0434\u0438\u0430\u043d\u044b \u0434\u043b\u044f K \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043b\u043e\u0441\u044c 64, \u0430 \u0432 \u043a\u0430\u0447\u0435\u0441\u0442\u0432\u0435 \u043c\u0430\u043a\u0441\u0438\u043c\u0443\u043c\u0430 - 178.<\/p>\n

\u041f\u0440\u043e\u0449\u0435 \u0433\u043e\u0432\u043e\u0440\u044f, F1@K - \u044d\u0442\u043e \u043f\u043e\u043a\u0430\u0437\u0430\u0442\u0435\u043b\u044c \"\u041f\u0440\u0435\u0434\u043e\u0441\u0442\u0430\u0432\u0438\u043b \u043b\u0438 \u043e\u0442\u0432\u0435\u0442 \u043c\u043d\u0435 \u0441\u0442\u043e\u043b\u044c\u043a\u043e \u0444\u0430\u043a\u0442\u043e\u0432, \u0441\u043a\u043e\u043b\u044c\u043a\u043e \u044f \u0445\u043e\u0442\u0435\u043b?\" \u0432 \u0441\u043e\u0447\u0435\u0442\u0430\u043d\u0438\u0438 \u0441 \"\u0421\u043a\u043e\u043b\u044c\u043a\u043e \u0438\u0437 \u044d\u0442\u0438\u0445 \u0444\u0430\u043a\u0442\u043e\u0432 \u043e\u043a\u0430\u0437\u0430\u043b\u0438\u0441\u044c \u043f\u0440\u0430\u0432\u0434\u0438\u0432\u044b\u043c\u0438?\".<\/p>\n

\u041a\u0430\u043a\u043e\u0439 LLM \u044f\u0432\u043b\u044f\u0435\u0442\u0441\u044f \u043d\u0430\u0438\u0431\u043e\u043b\u0435\u0435 \u0430\u043a\u0442\u0443\u0430\u043b\u044c\u043d\u044b\u043c?<\/h2>\n

\u0418\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u0442\u0435\u043b\u0438 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043b\u0438 LongFact, \u0447\u0442\u043e\u0431\u044b \u043f\u0440\u0435\u0434\u043b\u043e\u0436\u0438\u0442\u044c 13 LLM \u0438\u0437 \u0441\u0435\u043c\u0435\u0439\u0441\u0442\u0432 Gemini, GPT, Claude \u0438 PaLM-2. \u0417\u0430\u0442\u0435\u043c \u043e\u043d\u0438 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043b\u0438 SAFE \u0434\u043b\u044f \u043e\u0446\u0435\u043d\u043a\u0438 \u0444\u0430\u043a\u0442\u0438\u0447\u043d\u043e\u0441\u0442\u0438 \u0438\u0445 \u043e\u0442\u0432\u0435\u0442\u043e\u0432.<\/p>\n

GPT-4-Turbo \u0432\u043e\u0437\u0433\u043b\u0430\u0432\u043b\u044f\u0435\u0442 \u0441\u043f\u0438\u0441\u043e\u043a \u0441\u0430\u043c\u044b\u0445 \u0444\u0430\u043a\u0442\u043e\u0433\u0440\u0430\u0444\u0438\u0447\u043d\u044b\u0445 \u043c\u043e\u0434\u0435\u043b\u0435\u0439 \u043f\u0440\u0438 \u0433\u0435\u043d\u0435\u0440\u0430\u0446\u0438\u0438 \u0434\u043b\u0438\u043d\u043d\u044b\u0445 \u043e\u0442\u0432\u0435\u0442\u043e\u0432. \u0417\u0430 \u043d\u0435\u0439 \u0441\u043b\u0435\u0434\u0443\u044e\u0442 Gemini-Ultra \u0438 PaLM-2-L-IT-RLHF. \u0420\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\u044b \u043f\u043e\u043a\u0430\u0437\u0430\u043b\u0438, \u0447\u0442\u043e \u0431\u043e\u043b\u044c\u0448\u0438\u0435 LLM \u0431\u043e\u043b\u0435\u0435 \u0444\u0430\u043a\u0442\u043e\u0433\u0440\u0430\u0444\u0438\u0447\u043d\u044b, \u0447\u0435\u043c \u043c\u0430\u043b\u0435\u043d\u044c\u043a\u0438\u0435.<\/p>\n

\u0412\u044b\u0447\u0438\u0441\u043b\u0435\u043d\u0438\u044f F1@K, \u0432\u0435\u0440\u043e\u044f\u0442\u043d\u043e, \u043f\u043e\u0440\u0430\u0434\u0443\u044e\u0442 \u0441\u043f\u0435\u0446\u0438\u0430\u043b\u0438\u0441\u0442\u043e\u0432 \u043f\u043e \u043e\u0431\u0440\u0430\u0431\u043e\u0442\u043a\u0435 \u0434\u0430\u043d\u043d\u044b\u0445, \u043d\u043e \u0434\u043b\u044f \u043f\u0440\u043e\u0441\u0442\u043e\u0442\u044b \u043c\u044b \u043f\u0440\u0438\u0432\u043e\u0434\u0438\u043c \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\u044b \u044d\u0442\u0430\u043b\u043e\u043d\u043d\u044b\u0445 \u0440\u0430\u0441\u0447\u0435\u0442\u043e\u0432, \u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u043f\u043e\u043a\u0430\u0437\u044b\u0432\u0430\u044e\u0442, \u043d\u0430\u0441\u043a\u043e\u043b\u044c\u043a\u043e \u0442\u043e\u0447\u043d\u0430 \u043a\u0430\u0436\u0434\u0430\u044f \u043c\u043e\u0434\u0435\u043b\u044c \u043f\u0440\u0438 \u0432\u043e\u0437\u0432\u0440\u0430\u0449\u0435\u043d\u0438\u0438 \u043e\u0442\u0432\u0435\u0442\u043e\u0432 \u0441\u0440\u0435\u0434\u043d\u0435\u0439 \u0434\u043b\u0438\u043d\u044b \u0438 \u0431\u043e\u043b\u0435\u0435 \u0434\u043b\u0438\u043d\u043d\u044b\u0445 \u043e\u0442\u0432\u0435\u0442\u043e\u0432 \u043d\u0430 \u0432\u043e\u043f\u0440\u043e\u0441\u044b.<\/p>\n

\"\"
\u0414\u043b\u0438\u043d\u043d\u044b\u0435 \u043f\u043e\u043a\u0430\u0437\u0430\u0442\u0435\u043b\u0438 \u0444\u0430\u043a\u0442\u043e\u0433\u0440\u0430\u0444\u0438\u0447\u043d\u043e\u0441\u0442\u0438 13 LLM \u0441 K = 64 (\u043c\u0435\u0434\u0438\u0430\u043d\u043d\u043e\u0435 \u043a\u043e\u043b\u0438\u0447\u0435\u0441\u0442\u0432\u043e \u0444\u0430\u043a\u0442\u043e\u0432 \u0441\u0440\u0435\u0434\u0438 \u0432\u0441\u0435\u0445 \u043e\u0442\u0432\u0435\u0442\u043e\u0432 \u043c\u043e\u0434\u0435\u043b\u0438) \u0438 K = 178 (\u043c\u0430\u043a\u0441\u0438\u043c\u0430\u043b\u044c\u043d\u043e\u0435 \u043a\u043e\u043b\u0438\u0447\u0435\u0441\u0442\u0432\u043e \u0444\u0430\u043a\u0442\u043e\u0432 \u0441\u0440\u0435\u0434\u0438 \u0432\u0441\u0435\u0445 \u043e\u0442\u0432\u0435\u0442\u043e\u0432 \u043c\u043e\u0434\u0435\u043b\u0438). \u0418\u0441\u0442\u043e\u0447\u043d\u0438\u043a: arXiv<\/figcaption><\/figure>\n

SAFE - \u044d\u0442\u043e \u0434\u0435\u0448\u0435\u0432\u044b\u0439 \u0438 \u044d\u0444\u0444\u0435\u043a\u0442\u0438\u0432\u043d\u044b\u0439 \u0441\u043f\u043e\u0441\u043e\u0431 \u043a\u043e\u043b\u0438\u0447\u0435\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u0439 \u043e\u0446\u0435\u043d\u043a\u0438 \u0444\u0430\u043a\u0442\u043e\u043b\u043e\u0433\u0438\u0447\u043d\u043e\u0441\u0442\u0438 \u0434\u043b\u0438\u043d\u043d\u044b\u0445 \u0444\u043e\u0440\u043c LLM. \u041e\u043d \u0431\u044b\u0441\u0442\u0440\u0435\u0435 \u0438 \u0434\u0435\u0448\u0435\u0432\u043b\u0435, \u0447\u0435\u043c \u043b\u044e\u0434\u0438, \u043f\u0440\u043e\u0432\u0435\u0440\u044f\u0435\u0442 \u0444\u0430\u043a\u0442\u044b, \u043d\u043e \u0432\u0441\u0435 \u0440\u0430\u0432\u043d\u043e \u0437\u0430\u0432\u0438\u0441\u0438\u0442 \u043e\u0442 \u043f\u0440\u0430\u0432\u0434\u0438\u0432\u043e\u0441\u0442\u0438 \u0438\u043d\u0444\u043e\u0440\u043c\u0430\u0446\u0438\u0438, \u043a\u043e\u0442\u043e\u0440\u0443\u044e Google \u0432\u044b\u0434\u0430\u0435\u0442 \u0432 \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\u0430\u0445 \u043f\u043e\u0438\u0441\u043a\u0430.<\/p>\n

DeepMind \u0432\u044b\u043f\u0443\u0441\u0442\u0438\u043b\u0430 SAFE \u0434\u043b\u044f \u043f\u0443\u0431\u043b\u0438\u0447\u043d\u043e\u0433\u043e \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043d\u0438\u044f \u0438 \u043f\u0440\u0435\u0434\u043f\u043e\u043b\u043e\u0436\u0438\u043b\u0430, \u0447\u0442\u043e \u043e\u043d \u043c\u043e\u0436\u0435\u0442 \u043f\u043e\u043c\u043e\u0447\u044c \u0443\u043b\u0443\u0447\u0448\u0438\u0442\u044c \u0444\u0430\u043a\u0442\u043e\u043b\u043e\u0433\u0438\u0447\u043d\u043e\u0441\u0442\u044c LLM \u0437\u0430 \u0441\u0447\u0435\u0442 \u043b\u0443\u0447\u0448\u0435\u0433\u043e \u043f\u0440\u0435\u0434\u0432\u0430\u0440\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f \u0438 \u0442\u043e\u043d\u043a\u043e\u0439 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0438. \u041e\u043d\u0430 \u0442\u0430\u043a\u0436\u0435 \u043c\u043e\u0436\u0435\u0442 \u043f\u043e\u0437\u0432\u043e\u043b\u0438\u0442\u044c LLM \u043f\u0440\u043e\u0432\u0435\u0440\u044f\u0442\u044c \u0441\u0432\u043e\u0438 \u0444\u0430\u043a\u0442\u044b \u043f\u0435\u0440\u0435\u0434 \u0442\u0435\u043c, \u043a\u0430\u043a \u043f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u0438\u0442\u044c \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442 \u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u0442\u0435\u043b\u044e.<\/p>\n

OpenAI \u0431\u0443\u0434\u0435\u0442 \u0440\u0430\u0434 \u0443\u0432\u0438\u0434\u0435\u0442\u044c, \u0447\u0442\u043e \u0438\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u043d\u0438\u0435 Google \u043f\u043e\u043a\u0430\u0437\u0430\u043b\u043e, \u0447\u0442\u043e GPT-4 \u043f\u043e\u0431\u0435\u0436\u0434\u0430\u0435\u0442 Gemini \u0432 \u0435\u0449\u0435 \u043e\u0434\u043d\u043e\u043c \u0431\u0435\u043d\u0447\u043c\u0430\u0440\u043a\u0435.<\/p>","protected":false},"excerpt":{"rendered":"

\u0418\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u0442\u0435\u043b\u0438 \u0438\u0437 DeepMind \u0438 \u0421\u0442\u044d\u043d\u0444\u043e\u0440\u0434\u0441\u043a\u043e\u0433\u043e \u0443\u043d\u0438\u0432\u0435\u0440\u0441\u0438\u0442\u0435\u0442\u0430 \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u0430\u043b\u0438 \u0418\u0418-\u0430\u0433\u0435\u043d\u0442, \u043a\u043e\u0442\u043e\u0440\u044b\u0439 \u043f\u0440\u043e\u0432\u0435\u0440\u044f\u0435\u0442 \u0444\u0430\u043a\u0442\u043e\u043b\u043e\u0433\u0438\u044e LLM \u0438 \u043f\u043e\u0437\u0432\u043e\u043b\u044f\u0435\u0442 \u043f\u0440\u043e\u0432\u043e\u0434\u0438\u0442\u044c \u0441\u0440\u0430\u0432\u043d\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u0439 \u0430\u043d\u0430\u043b\u0438\u0437 \u0444\u0430\u043a\u0442\u043e\u043b\u043e\u0433\u0438\u0447\u043d\u043e\u0441\u0442\u0438 \u0418\u0418-\u043c\u043e\u0434\u0435\u043b\u0435\u0439. \u0414\u0430\u0436\u0435 \u0441\u0430\u043c\u044b\u0435 \u043b\u0443\u0447\u0448\u0438\u0435 \u043c\u043e\u0434\u0435\u043b\u0438 \u0418\u0418 \u0432\u0441\u0435 \u0440\u0430\u0432\u043d\u043e \u0438\u043d\u043e\u0433\u0434\u0430 \u0441\u043a\u043b\u043e\u043d\u043d\u044b \u043a \u0433\u0430\u043b\u043b\u044e\u0446\u0438\u043d\u0430\u0446\u0438\u044f\u043c. \u0415\u0441\u043b\u0438 \u0432\u044b \u043f\u043e\u043f\u0440\u043e\u0441\u0438\u0442\u0435 ChatGPT \u043f\u0440\u0435\u0434\u043e\u0441\u0442\u0430\u0432\u0438\u0442\u044c \u0432\u0430\u043c \u0444\u0430\u043a\u0442\u044b \u043f\u043e \u043a\u0430\u043a\u043e\u0439-\u043b\u0438\u0431\u043e \u0442\u0435\u043c\u0435, \u0442\u043e \u0447\u0435\u043c \u0434\u043b\u0438\u043d\u043d\u0435\u0435 \u0431\u0443\u0434\u0435\u0442 \u0435\u0433\u043e \u043e\u0442\u0432\u0435\u0442, \u0442\u0435\u043c \u0431\u043e\u043b\u044c\u0448\u0435 \u0432\u0435\u0440\u043e\u044f\u0442\u043d\u043e\u0441\u0442\u044c \u0442\u043e\u0433\u043e, \u0447\u0442\u043e \u043e\u043d \u0431\u0443\u0434\u0435\u0442 \u0441\u043e\u0434\u0435\u0440\u0436\u0430\u0442\u044c \u043d\u0435\u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u0444\u0430\u043a\u0442\u044b, \u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u043d\u0435 \u0441\u043e\u043e\u0442\u0432\u0435\u0442\u0441\u0442\u0432\u0443\u044e\u0442 \u0434\u0435\u0439\u0441\u0442\u0432\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u0438. \u041a\u0430\u043a\u0438\u0435 \u043c\u043e\u0434\u0435\u043b\u0438 \u0431\u043e\u043b\u0435\u0435 \u0442\u043e\u0447\u043d\u044b \u0432 \u0444\u0430\u043a\u0442\u0430\u0445, \u0447\u0435\u043c \u0434\u0440\u0443\u0433\u0438\u0435, \u043a\u043e\u0433\u0434\u0430 \u0433\u0435\u043d\u0435\u0440\u0438\u0440\u0443\u044e\u0442 \u0434\u043b\u0438\u043d\u043d\u044b\u0435 \u043e\u0442\u0432\u0435\u0442\u044b? \u0421\u043b\u043e\u0436\u043d\u043e \u0441\u043a\u0430\u0437\u0430\u0442\u044c, \u043f\u043e\u0442\u043e\u043c\u0443 \u0447\u0442\u043e \u0434\u043e \u0441\u0438\u0445 \u043f\u043e\u0440 \u0443 \u043d\u0430\u0441 \u043d\u0435 \u0431\u044b\u043b\u043e \u044d\u0442\u0430\u043b\u043e\u043d\u0430, \u0438\u0437\u043c\u0435\u0440\u044f\u044e\u0449\u0435\u0433\u043e \u0444\u0430\u043a\u0442\u043e\u043b\u043e\u0433\u0438\u0447\u043d\u043e\u0441\u0442\u044c \u0434\u043b\u0438\u043d\u043d\u044b\u0445 \u043e\u0442\u0432\u0435\u0442\u043e\u0432 LLM. \u0421\u043d\u0430\u0447\u0430\u043b\u0430 DeepMind \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043b\u0430 GPT-4 \u0434\u043b\u044f \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u044f LongFact, \u043d\u0430\u0431\u043e\u0440\u0430<\/p>","protected":false},"author":6,"featured_media":11182,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[147,118],"class_list":["post-11172","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-deepmind","tag-llms"],"yoast_head":"\nDeepMind developed SAFE, an AI agent to fact-check LLMs | DailyAI<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/dailyai.com\/ru\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/\" \/>\n<meta property=\"og:locale\" content=\"ru_RU\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"DeepMind developed SAFE, an AI agent to fact-check LLMs | DailyAI\" \/>\n<meta property=\"og:description\" content=\"Researchers from DeepMind and Stanford University developed an AI agent that fact-checks LLMs and enables benchmarking of AI model factuality. Even the best AI models still tend to hallucinate at times. If you ask ChatGPT to give you the facts about a topic, the longer its response the more likely it is to include some facts that aren\u2019t true. Which models are more factually accurate than others when generating longer answers? It\u2019s hard to say because until now, we didn\u2019t have a benchmark measuring the factuality of LLM long-form responses. DeepMind first used GPT-4 to create LongFact, a set of\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/ru\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2024-04-02T09:32:07+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/fact-vs-fake.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"1792\" \/>\n\t<meta property=\"og:image:height\" content=\"1024\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"author\" content=\"Eugene van der Watt\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:site\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:label1\" content=\"\u041d\u0430\u043f\u0438\u0441\u0430\u043d\u043e \u0430\u0432\u0442\u043e\u0440\u043e\u043c\" \/>\n\t<meta name=\"twitter:data1\" content=\"Eugene van der Watt\" \/>\n\t<meta name=\"twitter:label2\" content=\"\u041f\u0440\u0438\u043c\u0435\u0440\u043d\u043e\u0435 \u0432\u0440\u0435\u043c\u044f \u0434\u043b\u044f \u0447\u0442\u0435\u043d\u0438\u044f\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 \u043c\u0438\u043d\u0443\u0442\u044b\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/\"},\"author\":{\"name\":\"Eugene van der Watt\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\"},\"headline\":\"DeepMind developed SAFE, an AI agent to fact-check LLMs\",\"datePublished\":\"2024-04-02T09:32:07+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/\"},\"wordCount\":611,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/04\\\/fact-vs-fake.webp\",\"keywords\":[\"DeepMind\",\"LLMS\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"ru-RU\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/\",\"name\":\"DeepMind developed SAFE, an AI agent to fact-check LLMs | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/04\\\/fact-vs-fake.webp\",\"datePublished\":\"2024-04-02T09:32:07+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/#breadcrumb\"},\"inLanguage\":\"ru-RU\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"ru-RU\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/04\\\/fact-vs-fake.webp\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2024\\\/04\\\/fact-vs-fake.webp\",\"width\":1792,\"height\":1024},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2024\\\/04\\\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"DeepMind developed SAFE, an AI agent to fact-check LLMs\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"name\":\"DailyAI\",\"description\":\"Your Daily Dose of AI News\",\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/dailyai.com\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"ru-RU\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ru-RU\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"width\":4501,\"height\":934,\"caption\":\"DailyAI\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/DailyAIOfficial\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/dailyaiofficial\\\/\",\"https:\\\/\\\/www.youtube.com\\\/@DailyAIOfficial\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\",\"name\":\"Eugene van der Watt\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ru-RU\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"caption\":\"Eugene van der Watt\"},\"description\":\"Eugene comes from an electronic engineering background and loves all things tech. When he takes a break from consuming AI news you'll find him at the snooker table.\",\"sameAs\":[\"www.linkedin.com\\\/in\\\/eugene-van-der-watt-16828119\"],\"url\":\"https:\\\/\\\/dailyai.com\\\/ru\\\/author\\\/eugene\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"DeepMind \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u0430\u043b\u0430 SAFE, \u0418\u0418-\u0430\u0433\u0435\u043d\u0442\u0430 \u0434\u043b\u044f \u043f\u0440\u043e\u0432\u0435\u0440\u043a\u0438 \u0444\u0430\u043a\u0442\u043e\u0432 \u043e LLM | DailyAI","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/dailyai.com\/ru\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/","og_locale":"ru_RU","og_type":"article","og_title":"DeepMind developed SAFE, an AI agent to fact-check LLMs | DailyAI","og_description":"Researchers from DeepMind and Stanford University developed an AI agent that fact-checks LLMs and enables benchmarking of AI model factuality. Even the best AI models still tend to hallucinate at times. If you ask ChatGPT to give you the facts about a topic, the longer its response the more likely it is to include some facts that aren\u2019t true. Which models are more factually accurate than others when generating longer answers? It\u2019s hard to say because until now, we didn\u2019t have a benchmark measuring the factuality of LLM long-form responses. DeepMind first used GPT-4 to create LongFact, a set of","og_url":"https:\/\/dailyai.com\/ru\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/","og_site_name":"DailyAI","article_published_time":"2024-04-02T09:32:07+00:00","og_image":[{"width":1792,"height":1024,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/fact-vs-fake.webp","type":"image\/webp"}],"author":"Eugene van der Watt","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"\u041d\u0430\u043f\u0438\u0441\u0430\u043d\u043e \u0430\u0432\u0442\u043e\u0440\u043e\u043c":"Eugene van der Watt","\u041f\u0440\u0438\u043c\u0435\u0440\u043d\u043e\u0435 \u0432\u0440\u0435\u043c\u044f \u0434\u043b\u044f \u0447\u0442\u0435\u043d\u0438\u044f":"4 \u043c\u0438\u043d\u0443\u0442\u044b"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/"},"author":{"name":"Eugene van der Watt","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa"},"headline":"DeepMind developed SAFE, an AI agent to fact-check LLMs","datePublished":"2024-04-02T09:32:07+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/"},"wordCount":611,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/fact-vs-fake.webp","keywords":["DeepMind","LLMS"],"articleSection":["Industry"],"inLanguage":"ru-RU"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/","url":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/","name":"DeepMind \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u0430\u043b\u0430 SAFE, \u0418\u0418-\u0430\u0433\u0435\u043d\u0442\u0430 \u0434\u043b\u044f \u043f\u0440\u043e\u0432\u0435\u0440\u043a\u0438 \u0444\u0430\u043a\u0442\u043e\u0432 \u043e LLM | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/fact-vs-fake.webp","datePublished":"2024-04-02T09:32:07+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/#breadcrumb"},"inLanguage":"ru-RU","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/"]}]},{"@type":"ImageObject","inLanguage":"ru-RU","@id":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/fact-vs-fake.webp","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2024\/04\/fact-vs-fake.webp","width":1792,"height":1024},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2024\/04\/deepmind-developed-safe-an-ai-agent-to-fact-check-llms\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"DeepMind developed SAFE, an AI agent to fact-check LLMs"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"\u0412\u0430\u0448\u0430 \u0435\u0436\u0435\u0434\u043d\u0435\u0432\u043d\u0430\u044f \u0434\u043e\u0437\u0430 \u043d\u043e\u0432\u043e\u0441\u0442\u0435\u0439 \u043e\u0431 \u0438\u0441\u043a\u0443\u0441\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u043c \u0438\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0435","publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/dailyai.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"ru-RU"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"ru-RU","@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","width":4501,"height":934,"caption":"DailyAI"},"image":{"@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/DailyAIOfficial","https:\/\/www.linkedin.com\/company\/dailyaiofficial\/","https:\/\/www.youtube.com\/@DailyAIOfficial"]},{"@type":"Person","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa","name":"\u042e\u0434\u0436\u0438\u043d \u0432\u0430\u043d \u0434\u0435\u0440 \u0412\u0430\u0442\u0442","image":{"@type":"ImageObject","inLanguage":"ru-RU","@id":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","caption":"Eugene van der Watt"},"description":"\u042e\u0434\u0436\u0438\u043d - \u0432\u044b\u0445\u043e\u0434\u0435\u0446 \u0438\u0437 \u044d\u043b\u0435\u043a\u0442\u0440\u043e\u043d\u043d\u043e\u0439 \u0438\u043d\u0436\u0435\u043d\u0435\u0440\u0438\u0438 \u0438 \u043e\u0431\u043e\u0436\u0430\u0435\u0442 \u0432\u0441\u0435, \u0447\u0442\u043e \u0441\u0432\u044f\u0437\u0430\u043d\u043e \u0441 \u0442\u0435\u0445\u043d\u0438\u043a\u043e\u0439. \u041a\u043e\u0433\u0434\u0430 \u043e\u043d \u043e\u0442\u0434\u044b\u0445\u0430\u0435\u0442 \u043e\u0442 \u0447\u0442\u0435\u043d\u0438\u044f \u043d\u043e\u0432\u043e\u0441\u0442\u0435\u0439 \u043e\u0431 \u0438\u0441\u043a\u0443\u0441\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u043c \u0438\u043d\u0442\u0435\u043b\u043b\u0435\u043a\u0442\u0435, \u0432\u044b \u043c\u043e\u0436\u0435\u0442\u0435 \u043d\u0430\u0439\u0442\u0438 \u0435\u0433\u043e \u0437\u0430 \u0441\u0442\u043e\u043b\u043e\u043c \u0434\u043b\u044f \u0438\u0433\u0440\u044b \u0432 \u0441\u043d\u0443\u043a\u0435\u0440.","sameAs":["www.linkedin.com\/in\/eugene-van-der-watt-16828119"],"url":"https:\/\/dailyai.com\/ru\/author\/eugene\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/posts\/11172","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/comments?post=11172"}],"version-history":[{"count":2,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/posts\/11172\/revisions"}],"predecessor-version":[{"id":11181,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/posts\/11172\/revisions\/11181"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/media\/11182"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/media?parent=11172"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/categories?post=11172"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/ru\/wp-json\/wp\/v2\/tags?post=11172"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}