🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Fantastic!

The question that a reviewer should ask themselves is: Does this paper take a gradient step in a promising direction? Is the community better off with this paper published? If the answer is yes, then the recommendation should be to accept.

Both PnP and diffusion models (DMs) use denoisers. When comparing PnP to DM-based inverse problem solvers, one can use the same neural network as the prior. No need to use DnCNN or DRUNet. Thanks to @Ch2cago and Yuyang Hu✈️ICML for showing how PnP/RED with a DM network performs.

Netflix just dropped Go-with-the-Flow Motion-Controllable Video Diffusion Models Using Real-Time Warped Noise

Next stop arXiv cleaner github.com/google-researc…

Congrats, Albert! Thrilled he was the first undergrad I collaborated with. Wishing you great success at Caltech - keep in touch!

Gemma 3 explained: Longer context, image support, and a new 1B model. Read the blog for a deep dive into key enhancements → goo.gle/4lV8iaw

📄 “Stochastic Deep Restoration Priors for Imaging Inverse Problems” was accepted to ICML 2025 (ICML Conference)! Joint work with Yuyang Hu, Albert Peng, Weijie Gan, and colleagues from the Google’s Computational Imaging team, mauricio delbracio and Peyman Milanfar. wustl-cig.github.io/sharpwww/

Happy to announce that our work, ShaRP, has been accepted to #ICML2025! 🎉 Huge thanks to my collaborators Ulugbek S. Kamilov mauricio delbracio Peyman Milanfar Weijie Gan Albert Peng. Can't wait to see everyone in Vancouver!

Say goodbye to the silent era of video generation: Introducing Veo 3 — with native audio generation. 🗣️ Quality is up from Veo 2, and now you can add dialogue between characters, sound effects and background noise. Veo 3 is available now in the Google Gemini App for Google AI Ultra

Fantastic work by Matthieu! It's incredible to see how general restoration models (beyond denoisers) can act as such a strong prior for solving general inverse problems. Huge potential here!

“Random Walks with Tweedie: A Unified View of Score-Based Diffusion Models” has been accepted to IEEE Signal Processing Magazine (2025) IEEE SPM! Check out our concise theoretical justification for score-based diffusion models—and much more—here: wustl-cig.github.io/randomwalk/

Want to boost the performance of your pre-trained restoration diffusion model? Use KDE to steer your samples toward the mode during inference—no retraining needed. Check out our new work! 👉Link to the paper: [arxiv.org/abs/2507.05604].

Headed to ICML in Vancouver (July 12-17)! 🇨🇦 Catch me and Albert Peng on Tuesday, July 15, from 4:30 PM - 7:00 PM as we present our poster: "Stochastic Deep Restoration Priors for Imaging Inverse Problems." Let's chat about diffusion models, inverse problems, image/video