“Native image generation with Gemini 2.0 Flash in AI Studio is incredible.. 🔥 Can’t believe its free. Many paid Android apps doing this will feel the heat now.. https://x.com/rohanpaul_ai/status/1899936200682856714
“next. fucking. level https://x.com/multimodalart/status/1899887369811009802
“The new SVG Doodles is live. From using Handlebars to React with shadcn and Motion, built with Lovable. – Copy the SVG shape in SVG and TSX – Choose color – Copy and paste it anywhere, also in Figma. Will soon ad a search and a drop down to choose categories and new SVGs https://x.com/Mike_Andreuzza/status/1896858187128561904
“Gemini dropped their truly multimodal in/out image generator. Been on the EAP, here are my thoughts + some tips & tricks for “conversational image editing” Bookmark this, and try it later for free in google AI studio. Let’s get into it: https://x.com/bilawalsidhu/status/1899904526284710371
“completely sota for image editing! both generated and real images great 🚢 from google! i hope to see a gemma with image generation soon too ✨ https://x.com/multimodalart/status/1899881757396099231
“Super excited to ship Gemini’s native image generation into public experimental today 🙂 We’ve made a lot of progress and still have a way to go, please send us feedback! And yes, I made the image using Gemini. https://x.com/19kaushiks/status/1899856652666568732
“Luma Labs introduced a new pre-training technique called Inductive Moment Matching It produces superior image generation quality 10x more efficiently than current approaches Luma says the approach breaks the algorithmic ceiling of diffusion models! https://x.com/rowancheung/status/1899713364894392360
“Effective and Efficient Masked Image Generation Models “We propose a unified framework integrating masked image modeling and masked diffusion models” “on ImageNet 256Ă—256, with similar number of function evaluations (NFEs) and model parameters, eMIGM outperforms the seminal https://x.com/iScienceLuvr/status/1899411428236259603
“Inductive Moment Matching Luma AI introduces a new class of generative models for one- or few-step sampling with a single-stage training procedure. Surpasses diffusion models on ImageNet-256Ă—256 with 1.99 FID using only 8 inference steps and achieves state-of-the-art 2-step https://x.com/iScienceLuvr/status/1899409323651993976
[2503.04482] Generalized Interpolating Discrete Diffusion https://www.arxiv.org/abs/2503.04482
“Aligning Text to Image in Diffusion Models is Easier Than You Think “SoftREPA provides much improved text to image alignment by introducing a negligible size of learnable soft tokens.” https://x.com/iScienceLuvr/status/1899722047854404092
“Ray2 Keyframes, Extend & Loop is now here. Create your visual story with precise frame-by-frame control, seamless transitions, spatial exploration and long-form video durations using #Ray2 Text-to-Video and Image-to-Video models in Dream Machine. https://x.com/LumaLabsAI/status/1897334709514068451
AnyMoLe: Any Character Motion In-betweening Leveraging Video Diffusion Models https://kwanyun.github.io/AnyMoLe_page/
[2503.06923v1] From Reusing to Forecasting: Accelerating Diffusion Models with TaylorSeers https://arxiv.org/abs/2503.06923v1
“Today, we release Inductive Moment Matching (IMM): a new pre-training paradigm breaking the algorithmic ceiling of diffusion models. Higher sample quality. 10x more efficient. Single-stage, single network, stable training. Read more: https://x.com/LumaLabsAI/status/1899518379737661447
“Controlling Latent Diffusion Using Latent CLIP “In this paper, we introduce Latent-CLIP, a CLIP model that operates directly in the latent space. We train Latent-CLIP on 2.7B pairs of latent images and descriptive texts, and show that it matches zero-shot classification https://x.com/iScienceLuvr/status/1899715828419010769
[2503.06896v1] CATANet: Efficient Content-Aware Token Aggregation for Lightweight Image Super-Resolution https://arxiv.org/abs/2503.06896v1
“A rare impressive algorithmic breakthrough, seems like this just ends Consistency Models, maybe all of diffusion too. https://x.com/teortaxesTex/status/1899882749135106341
