Today in Generative Media
Meta's Llama 3.1 powers Meta AI; Condé Nast vs. Perplexity; AI takes video game jobs
News
Meta releases the biggest and best open-source AI model yet (The Verge)
Meta AI Is Now Multilingual, More Creative and Smarter (Meta blog)
Condé Nast has reportedly accused AI search startup Perplexity of plagiarism (Engadget)
AI Is Already Taking Jobs in the Video Game Industry (Wired)
Amazon’s paid Alexa is coming to fill a $25 billion hole dug by Echo devices (The Verge)
Mark Zuckerberg imagines content creators making AI clones of themselves (TechCrunch)
I finally feel grateful for AI, and it’s weirding me out (Fast Company)
Software
How Luma AI’s new ‘Loops’ feature in Dream Machine could transform digital marketing (VentureBeat)
Adobe's Photoshop can now generate AI images via prompts like Dall-E or Midjourney (Engadget)
Very, very fast voice bots. Llama 3.1 running on @GroqInc . 500ms voice-to-voice response times (X)
Research
Transparent Image Layer Diffusion using Latent Transparency (SIGGRAPH 2024, project page)
MovieDreamer: Hierarchical Generation for Coherent Long Visual Sequences (project page)
Masked Generative Video-to-Audio Transformers with Enhanced Synchronicity (project page)
A Simulation Benchmark for Autonomous Racing with Large-Scale Human Data (project page)
Computational Tradeoffs in Image Synthesis: Diffusion, Masked-Token, and Next-Token Prediction (arXiv)