Automated Design of Agentic Systems Presents Meta Agent Search to demonstrate that we can use agents to invent novel and powerful agent designs by programming in code proj: shengranhu.com/ADAS/ abs: arxiv.org/abs/2408.08435 github: github.com/ShengranHu/ADAS

JPEG-LM: LLMs as Image Generators with Canonical Codec Representations - Directly model images and videos via canonical codecs (e.g., JPEG, AVC/H.264) - More effective than pixel-based modeling and VQ baselines (yields a 31% reduction in FID) arxiv.org/abs/2408.08459

I can’t believe I finally met Lucidrains’ dog

Thanks a lot for joining our meetup! We had great turnout with so many interesting people showing up :) Some people even joined us just because they spotted a huge crowd we formed in Dolores Park they happened to walk across 😂

We released phi 3.5: mini+MoE+vision A better mini model with multilingual support: huggingface.co/microsoft/Phi-… A new MoE model:huggingface.co/microsoft/Phi-… A new vision model supporting multiple images: huggingface.co/microsoft/Phi-…

Join me and Aran Komatsuzaki tomorrow for our weekly AI papers/research space! Nous Research will join us to talk about their recent Hermes 3 release! x.com/i/spaces/1LyxB…

Come hear about building Hermes 3 with emozilla tomorrow 1PM PST on Tanishq Mathew Abraham, Ph.D. and Aran Komatsuzaki’s twitter space 🤗

Meta presents Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model - Can generate images and text on a par with similar scale diffusion models and language models - Compresses each image to just 16 patches arxiv.org/abs/2408.11039

WE ARE STARTING IN 6 MIN Hermes 3 - covered by emozilla from Nous Research A brief discussion on Phi 3.5 Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model JPEG-LM: LLMs as Image Generators with Canonical Codec Representations To Code, or

Reminder!! emozilla will be on Aran and Tanishq's spaces in 5 mins to talk about bulding Hermes with me and NobodyExistsOnTheInternet!

Over the coming days, start creating and chatting with Gems: customizable versions of Gemini that act as topic experts. 🤝 We’re also launching premade Gems for different scenarios - including Learning coach to break down complex topics and Coding partner to level up your skills

Your LLM may be sparser than you thought! Excited to announce TEAL, a simple training-free method that achieves up to 40-50% model-wide activation sparsity on Llama-2/3 and Mistral models. Combined with a custom kernel, we achieve end-to-end speedups of up to 1.53x-1.8x!

Improving 2D Feature Representations by 3D-Aware Fine-Tuning proj: ywyue.github.io/FiT3D/ abs: arxiv.org/abs/2407.20229

AI2 presents OLMoE: Open Mixture-of-Experts Language Models - Opensources SotA LMs w/ MoE up to 7B active params. - Releases model weights, training data, code, and logs. repo: github.com/allenai/OLMoE hf: huggingface.co/allenai/OLMoE-… abs: arxiv.org/abs/2409.02060

Join me and Aran Komatsuzaki today for our AI papers of the week space! x.com/i/spaces/1OyJA…

STARTING IN 10 MIN!!! Papers we will cover: Building and better understanding vision-language models: insights and future directions - presented by Leo Tronchon OLMoE: Open Mixture-of-Experts Language Models - presented by Niklas Muennighoff Diffusion Models Are Real-Time Game

Yoshua Bengio started using X x.com/Yoshua_Bengio

MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark - Tests a cognitive skill of seamlessly integrating visual and textual information - Performance is substantially lower than on MMMU, ranging from 16.8% to 26.9% across models proj:

We’re presenting AlphaProteo: an AI system for designing novel proteins that bind more successfully to target molecules. 🧬 It could help scientists better understand how biological systems function, save time in research, advance drug design and more. 🧵 dpmd.ai/3XuMqbX