Same here 😄

Honored that our paper won Best Paper at the CVPR VisCon workshop! I wish I could have been there. Big thanks to my coauthors and the organisers.

Submit your papers to P13N workshop at ICCV 2025. It will be fun. Talks are amazing, panelists are awesome. And, I will be there 😄 #ICCV2025

ICCV decisions are out — if your paper didn’t make it, don’t worry! Submit your work to the P13N Workshop instead! Let’s push the frontier of personalized generative AI together!💡 #ICCV2025 #P13NWorkshop #Personalization #ICCV2025 More info: p13n-workshop.github.io

Really excited for the Personalization in Generative AI Workshop at #ICCV2025 Sumit by July 7 (long papers); Aug 18 (short papers) & join us with amazing talks at Hawaii!! 🔗Submit: openreview.net/group?id=thecv… 🌐Info: p13n-workshop.github.io

The first Personalization in GenAI Workshop is coming to #ICCV2025 in Hawaii! World-class speakers & panelists Cutting-edge research showcase Network with industry leaders Don't miss your chance to be part of this groundbreaking event! ⌛️Deadline: July 07 - Submit

Excited that our work, Concept Attention, was selected as an Oral Presentation at ICML 2025 and recently won the Best Paper Award at the CVPR Workshop on Visual Concepts. Concept Attention allows you to visualize the presence of text concepts in generated videos and images.

Our work ConceptAttention has been selected as an Oral Presentation at ICML. Alec will be there presenting, go check it out if you’re attending!

ConceptAttention: Diffusion Transformers Learn Highly Interpretable Features Oral presentation Alec Helbling Tuna Meral (VT) Ben Hoover Pinar Yanardag (VT) Duen Horng Chau arxiv.org/abs/2502.04320 TL;DR: Saliency without supervision: unlocks hidden concept awareness in DiTs

NeurIPS Conference, why take the option to provide figures in the rebuttals away from the authors during the rebuttal period? Grounding the discussion in hard evidential data (like plots) makes resolving disagreements much easier for both the authors and the reviewers. Left: NeurIPS

The best part of the new NeurIPS Conference rebuttal policy is getting to write a 500-word essay describing what my image generation output for "a Cthulhu-themed waterpark" looks like for Reviewer 2 instead of just... showing them the picture. 😅

I can’t stop mentioning about fal to people. Cool people, congrats 🎊

Who will watch the watchmen? 😅 github.com/TencentQQGYLab…

My favorite part of reviewing “6” papers for a single conference is the email telling us to be “fair” right before threatening our future publications. This delightful mix of professional terror and demand for free labor is an exemplar ✨ motivational leadership ✨

Ph.D. student Tuna Meral's (Tuna Meral) home base for summer was Amazon Science AGI Foundations, San Francisco. As applied scientist intern on video generation team, he helped develop autoregressive models for creating both images & video. His advisor The Sanghani Center at Virginia Tech: Pinar Yanardag

Worked on really cool stuff 🚀🚀🚀

What next?