Zeitpunkt Nutzer Delta Tröts TNR Titel Version maxTL Di 30.07.2024 00:00:41 73.447 0 2.423.669 33,0 Mastodon 4.2.10 500 Mo 29.07.2024 00:00:04 73.447 -1 2.421.080 33,0 Mastodon 4.2.10 500 So 28.07.2024 00:00:14 73.448 -2 2.418.511 32,9 Mastodon 4.2.10 500 Sa 27.07.2024 00:00:07 73.450 +1 2.416.075 32,9 Mastodon 4.2.10 500 Fr 26.07.2024 00:00:06 73.449 -1 2.413.147 32,9 Mastodon 4.2.10 500 Do 25.07.2024 00:00:07 73.450 0 2.411.145 32,8 Mastodon 4.2.10 500 Mi 24.07.2024 00:00:55 73.450 0 2.407.498 32,8 Mastodon 4.2.10 500 Di 23.07.2024 00:00:35 73.450 0 2.404.000 32,7 Mastodon 4.2.10 500 Mo 22.07.2024 00:02:39 73.450 0 2.400.400 32,7 Mastodon 4.2.10 500 So 21.07.2024 00:00:19 73.450 0 2.397.821 32,6 Mastodon 4.2.10 500
Bornach (@bornach) · 11/2022 · Tröts: 2.959 · Folger: 181
Di 30.07.2024 10:02
AI scientist's shocking revelation in this [Machine Learning Street Talk] interview that #ChatGPT doesn't reason
https://youtu.be/y1WnHpedi2A
Next week, AI scientists will tell us what bears do in the woods
Behind the clickbait title, Prof. Subbarao Kambhampati describes how researchers can show that LLMs are just really good at memorizing answers and are not performing even basic reasoning tasks
#LargeLanguageModels
#LLM #ChatGPT #GPT4 #AGI
[Öffentlich] Antw.: 0 Wtrl.: 0 Fav.: 0 · via Web