Il y a 41 ans jour pour jour naît le MacIntosh Ce 24 janvier 1984, le monde entre dans la révolution cognitive En 1994 Internet s’ouvre au grand public Google arrive en 1998 2011, les réseaux de neurones artificiels se démocratisent En 2025 l’IA dépasse notre cerveau
MACHINE LEARNING
-
Reasoning on obscure verifiable academic topics implies data-efficient learning
By
–
Reasoning well in general at the frontier of obscure but verifiable academic topics implies data-efficient learning And it’s not quite memorization — by design HLE excludes questions non-experts could answer via Google. I think the tendon count here is meant to be newly deduced
-
Nobel laureate challenges AI skeptics on autocomplete misconception
By
–
Cc all AI-Skeptics who are still on the ‘fancy autocomplete’ bandwagon. (I would take this Noble Prize winner seriously when he says this. ) https://t.co/uWuvqOIO0L
— Nina Schick (@NinaDSchick) 24 janvier 2025Cc all AI-Skeptics who are still on the ‘fancy autocomplete’ bandwagon. (I would take this Noble Prize winner seriously when he says this. )
-
AI in Neurosurgical Practices: Current Trends and Future Opportunities
By
–
AI in Neurosurgical Practices: Current Trends and Future Opportunities
#AI #AIio #BigData #ML #NLU #Futureofwork @CRudinschi @AntonioSelas @alexjc @RobotLaunch @karpathy @andyjankowski @bobgourley @CadeMetz -
Strengthen Chatbot Predictive Models for Better User Understanding
By
–
Strengthen those chatbot predictive models to make sure every user is understood! Ready to begin? Register for this upcoming demo and Q&A session: https://
snorkel.ai/webinar/weekly
-demo-with-qa/
… #SnorkelAI #Chatbot -
AI Show Doubles Episode Coverage for Weekly News
By
–
Next week of The Artificial Intelligence Show (@aishowpod) will be 2 episodes. No way to cover all the AI news from this week in 1 episode. pic.twitter.com/1iP77qN0Q2
— Paul Roetzer (@paulroetzer) 23 janvier 2025Next week of The Artificial Intelligence Show (
@aishowpod
) will be 2 episodes. No way to cover all the AI news from this week in 1 episode. -
Humanity’s Last Exam: No AI model scores above 10%
By
–
Examples from Humanity’s Last Exam, designed by CAIS & Scale to be the *final* broad-subject, closed-end Q&A exam for LLMs written by humans. Today, no model scores above 10%. This is how hard a test has to be for frontier AI to score that low — in Classics, Ecology, and C.S.:
-
AI Progress on Packaged Tasks vs Real-World Jobs
By
–
It’s done because it’s much easier to 1) collect, 2) evaluate, and 3) beat and make progress on. We’re going to see every task that is served neatly packaged on a platter like this improved (including those that need PhD-grade expertise). But jobs (even intern-level) that need
-
More Bits Enable Greater Precision in Larger AI Models
By
–
more bits = more decimal places = more precision = bigger model
-
GenAI Models Molecular Motion for Chemistry and Biology
By
–
GenAI has shown potential in helping chemists & biologists explore static molecules, like proteins & DNA. One challenge, though: Molecules are constantly moving & jiggling, which is important to model for some proteins & drugs. Simulating these motions on a computer using