RT @davidbau: This Friday NEMI 2025 is at Northeastern in Boston, 8 talks, 24 roundtables, 90 posters; 200+ attendees. Thanks to @GoodfireA….

RT @runwayml: Introducing Runway Aleph, a new way to edit, transform and generate video. Aleph is a state-of-the-art in-context video mode….

Join millions who have switched to Grok.

RT @runwayml: Early access for a new kind of generative story experience with Game Worlds. Using References to turn a single image into a m….

RT @yukyunglee_: Can coding agents autonomously implement AI research extensions?. We introduce RExBench, a benchmark that tests if a codin….

On turning one as a faculty:

RT @dutran: Every year at @CVPRConf (and other conferences), a lot of money was spent on printing these posters (~150$ x 2500=375k), and af….

With my students at #CVPR2025 Tianle, @xavierohan and @dahye_kim 💪

On the importance of building a welcoming research community and doing meaningful, collaborative research by @sainingxie.

I’ll be talking today at 10:30 about existing challenges and steps for better evaluation of generative media. See you all there!.

Come meet me and my students at #CVPR25 Below is a list of works on interpretability and controllability of generative models that my students will be presenting🪩. 1. Revelio: Interpreting and leveraging semantic information in diffusion models 2.

RT @ma_nanye: Join us for a full-day tutorial on Scalable Generative Models in Computer Vision at @CVPR in Nashville, on Wednesday, June 11….

RT @jbhuang0604: How to present a poster at a conference?. Poster presentation is a great way to disseminate your research findings, have i….

RT @graceluo_: ✨New preprint: Dual-Process Image Generation! We distill *feedback from a VLM* into *feed-forward image generation*, at infe….

RT @runwayml: Join us on Thursday June 12th for our annual CVPR Friends Dinner. RSVP at the link below.

RT @c_valenzuelab: CVPR returns this June. Join us Thursday June 12th for our annual CVPR Friends Dinner. RSVP at the link below. https://t….

Wrapped up teaching Applied Machine Learning this semester🎇🎇. It was all-consuming, costed many sleepless nights, and a fair bit of venting. FWIW, students seemed to have enjoyed the class. Slides and video recordings available here:

[2/2] Our simple trick yields higher visual quality, better prompt alignment, 83% improvement in win-rate, and 8% higher VQA than baseline -- all at no increase in the inference cost. Website: .Work done by @KetanSuhaasS and @xavierohan 👏 .Presenting.

[1/2] 🔥 We present ScoPE, a training-free approach to adapt any diffusion model to effectively capture very long and detailed prompts. Inspired by human visual system, we force the model to first generate broad scene strokes then fill in the fine-grained details.

RT @WiCVworkshop: ✈️🎉 Travel support opportunities for the WiCV Workshop at @CVPR 2025 are open!. 1⃣ WiCV Travel Grant: .