Our recent finding on Diffusion Alignment: a reward model in pixel space can be easily transferred to score noisy diffusion latents directly — at small finetuning cost, via stitching. This makes Faster & Better for both Training & Inference Alignment. Meet StitchVM👇 1/

Want to leverage the power of SOTA 3D models like VGGT & Video LDMs for 3D generation? Now you can! 🚀 Introducing VIST3A — we stitch pretrained video generators to 3D foundation models and align them via reward finetuning. 📄 arxiv.org/abs/2510.13454 🌐 gohyojun15.github.io/VIST3A

🚀 Just released: FLAIR – a new training-free approach to solving inverse problems using flow-matching models! 🎯 Try it live: huggingface.co/spaces/prs-et… 📚 Learn more: inverseflair.github.io/

It works surprisingly well for memes with text:

🚀 Excited to Share: Solving Inverse Problems with FLAIR! 🚀 We present FLAIR, a novel framework for inverse problems using generative flow-based models. Project Page 🚀 : inverseflair.github.io arXiv 📜 : arxiv.org/abs/2506.02680 Demo 🤗 : huggingface.co/spaces/prs-et…

RollingDepth rolls into Nashville for #CVPR2025! 🎸

We present Thera🔥: The new SOTA arbitrary-scale super-resolution method with built-in anti-aliasing. Our approach introduces Neural Heat Fields, which guarantee exact Gaussian filtering at any scale, enabling continuous image reconstruction without extra computational cost.

A Variational Perspective on Generative Protein Fitness Optimization Uses an approach to optimize protein fitness in latent space. Famous AAV dataset by Bryant used. P: arxiv.org/abs/2501.19200

Introducing 🛹 RollingDepth 🛹 — a universal monocular depth estimator for arbitrarily long videos! Our paper, “Video Depth without Video Models,” delivers exactly that, setting new standards in temporal consistency. Check out more details in the thread 🧵