Mom of 3. Assistant Prof. #VirginiaTech CS. Formerly #MIT #Purdue #Bogazici. #Emmy Nominated Creative Director, #Fulbright Fellow. Contact: [email protected].
United States
Joined July 2007
- Tweets1.4K
- Following918
- Followers5K
- Likes1.4K
Pinned Tweet
Thrilled to receive an NSF CAREER Award! Excited to push the frontiers of video generation toward controllable, cinematic tools for creating dynamic worlds. Grateful to my students, mentors, collaborators, and NSF. nsf.gov/awardsearch/show-awa…
Pinar Yanardag retweeted
SPAWN got accepted to #NeurIPS2026 🎉 In this work, we study controllability in interactive world models by injecting custom concepts.
Many thanks to my advisor @PINguAR for the guidance and support, and @fal for providing the compute resources that made this project possible!
Pinar Yanardag retweeted
Huge thanks to the professors at @SanghaniCtrVT! Dr. @jiamingcui1997 #muratkantarcioglu @PINguAR and @Ismini_L for their invaluable guidance and support in realizing these projects. Many thanks to my collaborators @yusuf_dalva @phylo_GENETIC #saadamin!
Pinar Yanardag retweeted
AdaState is accepted to #NeurIPS2026 🎉
I'm deeply grateful to my advisor @PINguAR for giving me the freedom to pursue this idea and for her support throughout!
Thanks also to the Amazon–Virginia Tech Initiative (@AmazonScience, @virginia_tech) for supporting this work, and to @fal for the compute support. Congrats to @kavanav2912 on the acceptance of RAVEL too!
As I approach the end of my PhD, I feel fortunate to have presented at CVPR, SIGGRAPH, ECCV, and NeurIPS as an author, all this year. None of it would have been possible without my amazing collaborators, and I'm thankful to every one of them.
See you at NeurIPS (probably Atlanta 🫠)
RT @d_yesiltepe: Super excited to share that VideoMLA has been accepted at #NeurIPS2026!
We studied the transition from Multi-head Self-At…
This quoted post is unavailable.
5 papers from our lab accepted at #NeurIPS2026! 🎉 So proud of our team! Our work spans video generation (AdaState, VideoMLA), controllable world models (SPAWN), and physics-aligned video generation (HiPhy). See more details at: gemlab-vt.github.io
🤖 Made with AI
Huge kudos to all the students and collaborators behind these papers! So proud of what we’ve accomplished together, and grateful to work with such a wonderful team! 💎 @d_yesiltepe @yusuf_dalva @TKazimi415 @akdemir_kiymet @tunahansalih @kavanav2912 @kaan_oktay @akaan_akan
GEMLAB is at #ECCV with two papers! 🎉 Find Yusuf (@yusuf_dalva) presenting at poster #125 and Ritika (@ritika1729) at poster #166. Stop by to check out our work and say hello!
Ritika @ritika1729 is presenting her paper on interpretable and diverse semantics for t2i models!
Pinar Yanardag retweeted
Enter Sandman is a 25 year old Virginia Tech tradition and the reason Metallica added the school to their last tour.
Virginia Tech’s Geosciences department measured the equivalent of a magnitude 1 to 2 earthquake when the band played the song at that concert.
Tried the demo with @Ismini_L a few months ago, and it was absolutely AMAZING! 🔥
1/ We’re entering the era of agentic commerce, and we believe that world models are the missing engine.
Instead of browsing static product pages, agents will let you interact with products inside your own world before making decisions.
Today, @DecartAI we're releasing Anywear👇
Go Rana! 🚀 So exciting to see Thrixel in action, very impressive results!
Our NeurIPS rebuttals is currently peer-reviewed by this tumbleweed. #NeurIPS @NeurIPSConf
Everything is automated by AI now, but waiting for NeurIPS reviews still feels beautifully old-school: repeatedly refreshing a webpage and using paper IDs to estimate how close the reviews are! Good luck, everyone!🤞#NeurIPS @NeurIPSConf
If you are at #SIGGRAPH2026, please check out my PhD student @yusuf_dalva's talk at 9:40AM today!
Happening this morning at #SIGGRAPH2026! I’m presenting Canvas-to-Image in the Image Generation session (9–10:30am PDT, Room 408 B), my talk’s at 9:40. Come through if you’re into multimodal / compositional control.
What an amazing crowd at the #ICML F2S Workshop today! 🔥 It was a blast sharing how we can unlock training-free control spanning from video generation to world models. The room was packed with great questions and exciting discussions. Huge thanks to the incredible organizers for making this workshop happen, and to everyone who joined! 🚀 #ICML
I took Gilbert Strang’s course at MIT, and it profoundly shaped how I think. One fun memory: an acappella group stormed into class on Valentine’s Day to serenade a student, Prof Strang was thoroughly amused! @MIT_CSAIL
IN 1999 MIT FILMED A MATH LECTURE THAT QUIETLY BECAME THE FOUNDATION OF EVERY AI MODEL YOU'VE EVER USED AND ALMOST NO ONE WAS TAUGHT TO SEE IT THAT WAY
39 minutes from Gilbert Strang, who taught this at MIT for over 60 years -- the linear algebra course an entire generation of engineers and data scientists grew up on.
-> The shift it creates: you stop seeing matrices as boring grids of numbers and start seeing them as the language of space, data, and motion itself.
School drilled you to crunch matrices by hand and never told you why. Strang shows you what they actually mean.
Every neural net, every embedding, every model you prompt is linear algebra running underneath. The math you skipped is the engine of the thing you use all day.
Memorizing the steps was never the skill -> seeing what the numbers do is. This is where it finally clicks.
Most people fear linear algebra and move on. The ones who watched this see straight into how AI actually works.
Bookmark & Watch it today, this one's a legend ↓
We need a clear policy for “ghosting after rebuttal.” A review should be heavily discounted—or excluded—when the reviewer does not engage, unless the AC independently verifies that the concerns remain valid.