Vous Γͺtes en panne de lecture pendant les #JeuxOlympiques ? πππ€ π½πππ€ππ€ ππ ππ π€ππππππ sont lΓ pour vous ! Ne ratez pas la derniΓ¨re Γ©dition de la saison avec une avalanche d'annonces autour de l'#IA : -La sortie de #Llama3 de #Meta et un Mark Zuckerberg
GENERATIVE AI
-
Nvidia launches NIMs to accelerate generative AI applications
By
–
Nvidia announces raft of 'NIMs' to speed up Gen AI apps: The chip giant also announced a partnership with Hugging Face for inference-as-a-service. https://
zdnet.com/article/nvidia
-announces-raft-of-nims-to-speed-up-gen-ai-apps/?utm_source=dlvr.it&utm_medium=twitter#ftag=RSSbaffb68
β¦ #BigData #DataScience #IoT -
Microsoft’s Different Copilots: Features and Uses
By
–
What are Microsoft's different Copilots? Here's what they are and how you can use them: Microsoft has a series of AI companions with different functions that can help everyone, from business professionals to everyday users. https://
zdnet.com/article/what-i
s-microsoft-copilot-heres-everything-you-need-to-know/?utm_source=dlvr.it&utm_medium=twitter#ftag=RSSbaffb68
β¦ #BigData #DataScience #IoT -
ChatGPT GPT-4o Feature Digitizes Handwritten Notes Free
By
–
How to use ChatGPT to digitize your handwritten notes for free: This underrated GPT-4o feature can transform your handwritten lists and other documents into searchable content you can store in the cloud. https://
zdnet.com/article/how-to
-use-chatgpt-to-digitize-your-notes-for-free/?utm_source=dlvr.it&utm_medium=twitter#ftag=RSSbaffb68
β¦ #BigData #DataScience #IoT -
Large AI Model Shows Redundancy in Weight Parameters
By
–
It looks pretty good but expected even better from such a big model (so much redundancy in these weights)
-
GPT Number Comparison Ability as AGI Threshold Indicator
By
–
when you create GPTs that can accurately compare 9.9 and 9.11, AGI will be achieved.
-
Segment Anything 2: Real-Time Video Segmentation Innovation
By
–
Meta recently released Segment Anything 2, the next iteration in the SAM family which extends the segmentation capability to videos in real-time. https://t.co/LlzMHPmaJN
— Satya Mallick (@LearnOpenCV) 30 juillet 2024
In our article, we dive into the innovative approach of SAM 2, discussing its architecture, the data engine,β¦ pic.twitter.com/gFwOq6LNwPMeta recently released Segment Anything 2, the next iteration in the SAM family which extends the segmentation capability to videos in real-time. https://
learnopencv.com/sam-2/ In our article, we dive into the innovative approach of SAM 2, discussing its architecture, the data engine, -
AI Enhancing Human Creativity Empathy and Social Interactions
By
–
In this video, discover how #AI is challenging us to become more human by enhancing our creativity, empathy, and complex social interactions. Learn about the synergy between #AI and #humanintelligence and the evolving roles in various sectors.
-
LongGPT: 64k Tokens for $18
By
–
BREAKING: LongGPT is coming out! > 64k tokens = 50.000+ words.
> $18 for 1 Million tokens
> You can now generate 300-page books in minutes! -
LLMs struggle with spatial understanding but improvement expected soon
By
–
In the future definitely yes. But right now LLMs are not that great at spatial understanding, so they would be very confused. But this will get solved soon.