๐ Imagine an AI with a photographic memory of your conversations and lightning-fast text processing. With up to 22x faster inference, it preserves context effortlessly. No more forgotten chats or text limitations. StreamingLLM redefines language models! ๐ฌโจ
๐ Paper: https://t.co/1OccOgqe1n
#AI #StreamingLLM #Innovation
๐ค We are thrilled to announce our partnership with @2CRSi and @NVIDIA, two undisputed leaders in the IT and Artificial Intelligence industry.
This partnership will take our research to the next level!
#DataScience#AI#Partnership ๐
๐ IBM and NASA open source largest geospatial AI foundation model on Hugging Face! This collaboration will democratize access to AI, allowing scientists and researchers to analyze large environmental datasets and drive innovations in climate and Earth science. The potential to track deforestation, predict crop yields, and monitor greenhouse gases is enormous!
This breakthrough will accelerate all 3D developments! ๐ Imagine what more it could achieve with additional prompts! The possibilities are endless! ๐
๐ฃ Big news today! ๐ฃ
RosIA is now officially registered in the National Directory of Associations (RNA). It is with immense pride that we announce this crucial milestone achieved successfully!
We would like to express our gratitude to Danielle COLLA RIVIERE for her assistance.
Another Meta open-source projetc: AudioCraft.
A powerful family of generative AI models designed to create realistic audio & music from text. ๐ต With MusicGen, AudioGen, and EnCodec in one code base, it's now easier to compose music, generate sounds, and compress audio. ๐ง This release comes with an improved EnCodec for higher quality music generation and pretrained AudioGen models for environmental sounds and effects. ๐ณ๐
GitHub repository: https://t.co/LjfFdr64ji
๐ Curious about "Interpretability and Explainability in Machine Learning"? Discover the power of SHAP! ๐ Unveil intriguing patterns like gender influencing survival chances and age playing a role. Gain transparency and trust with #Interpretability and #Explainability in #MachineLearning. ๐กโจ
Interested? Check our blog article ๐ https://t.co/9IJ2UAaW9O
Sounds like great news! Introducing LP-MusicCaps ๐ถ Researchers leverage Large Language Models to generate natural language descriptions for music tracks. The dataset contains 2.2M captions and 0.5M audio clips, boosting music captioning performance and surpassing supervised baselines. ๐ต
AI's potential to revive ancient molecules for modern drug discovery is fascinating! While the current results may be promising, further improvements in the algorithm could lead to groundbreaking advancements in antibiotic development. Exciting times ahead for bioengineering and medicine! ๐งฌ๐ฌ
I've worked with superconductors for the better part of a decade now in different contexts, from STM condensed matter labs, to particle accelerators, and now fusion.
Time for a deep dive on what exactly this miracle-technology unlocks for us a species: ๐งต
๐ Exciting news from @StabilityAI! Presenting Stable Difusion XL 1.0 (SDXL), their latest model!
We put it to the test against the top contender, @midjourney 5.2, using the prompt "A pink astronaut climbing up a mountain", and the outcome is in! ๐๏ธ๐จโ๐
Which is your favourite?
Struggling with code issues? ๐ค Who hasn't used #ChatGPT to solve coding problems? But hey, 1 in 4 times, it throws errors itself! ๐
OverflowAI Search sounds like a game-changer, offering instant and reliable problem-solving using GenAI technology. Can't wait to try it out and find even better solutions to coding challenges! ๐๐ก
The net-net:
No champagne yet, but watch closely - this would be a serious game changer in things like power transmission, energy storage, and future-tech like quantum computers, fusion energy, mag-lev trains.
I'm even more optimistic than 6 weeks ago
https://t.co/eDpvsCRghH
After taking part in several challenges, Iโd highly recommend giving it a try if youโre a DS and want to take you first steps into Web3 + Learn about new subjects! ๐
Thanks for the talk ๐ค, canโt wait to see how the models created can be used for business purposes via dApp
๐ Finally, LLaMA 2 by @MetaAI is here!
โ Open-source license allowing commercial use
๐ Trained on 2 billion tokens
๐ 4096 context window (expandable)
๐งฎ Model versions with 7B, 13B, and 70B parameter
๐ง 70B model adopts grouped-query attention (GQA)
๐ Available on @Azure, @awscloud, and @huggingface
๐ Outperforms open-source models
๐ช Comparable performance to OpenAI's ChatGPT.
We can't wait to start using this new model from Meta AI at Aqsone!
๐ To contribute to the Data Science and Archaeological community, we are sharing our Vesuvius Challenge - Ink Detection solution with you.
Here is a link to our GitHub project: https://t.co/d8bJHCeRHR
๐ท Introducing ControlNet 1.1
Controls image generation with an image input!
Based on Stable Diffusion.
๐ See our logo transformed by ControlNet 1.1
๐ฌ Try it here: https://t.co/pjvJINYTyy
#AI#ImageProcessing#StableDiffusion#ControlNet