GEMLAB at CVPR 2026
Meet GEMLAB at CVPR.
A compact guide to GEMLAB papers, workshop presentations, invited talks, and panels at CVPR 2026, spanning controllable video generation, image editing, personalization and fairness.
01CVPR & Findings papers
Papers at CVPR 2026
Four GEMLAB papers across the main conference and the Findings track.
Diverse Video Generation with Determinantal Point Process-Guided Policy Optimization
DTG-Restore: Training-Free Diffusion Refinement for Generative Video Super-Resolution
LayerFusion: Harmonized Multi-Layer Text-to-Image Generation with Generative Priors
02Workshop presentations
Workshops & posters
Workshop papers and poster presentations across AI4CC, WiCV, and the AC Workshop.
MIST: Debiasing Pre-trained T2I Models with Cross-Attention Steering via Global-Token Alignment
Cognitive Canvas: Cognitive Enhancement through Text-to-Image Diffusion Models
Model-Agnostic and Training-Free Discovery of Semantic Directions for Text-to-Image Editing
AC Workshop Poster Presentation
03Invited talks & panels
PI talks & panels
Keynotes and a panel appearance across four CVPR 2026 workshops.
AI for Visual Arts
AI for Visual Arts
AI-assisted Long Video Creation Workshop
Personalization in Generative AI Workshop
04Where to find us
At-a-glance schedule
Every GEMLAB appearance, day by day — June 2 to June 7, 2026.
Jun 2
Tuesday
AC Workshop poster presentationRoom 601/603
Jun 3
Wednesday
AI4CC workshop papersRoom 610
WiCV poster sessionRoom 708
AI for Visual Arts keynote and panelRoom 4AB
AI-assisted Long Video Creation keynoteRoom 712
Jun 4
Thursday
Personalization in Generative AI keynoteRoom 4CD
Jun 5
Friday
Diverse Video Generation posterPoster Session 2 & Exhibit Hall
Jun 6
Saturday
DTG-Restore posterPoster Session 4 & Exhibit Hall
Jun 7
Sunday
LayerFusion Findings posterExHall A - Findings · 7:30–9:00 AM
∞-RoPE posterPoster Session 6 · 17:30
05Track record
GEMLAB at previous CVPRs
A look back at GEMLAB's CVPR 2025 and 2024 papers.
CVPR 20253 papers
- Yusuf Dalva, Kavana Venkatesh, and Pinar Yanardag
- Enis Simsar, Thomas Hofmann, Federico Tombari, and Pinar Yanardag
- Tahira Kazimi*, Ritika Allada*, and Pinar Yanardag
CVPR 20243 papers
- Yusuf Dalva and Pinar Yanardag
- Tuna Han Salih Meral, Enis Simsar, Federico Tombari, and Pinar Yanardag
-
HighlightRAVE: Randomized Noise Shuffling for Fast and Consistent Video Editing with Diffusion Models ↗Ozgur Kara*, Bariscan Kurtkaya*, Hidir Yesiltepe, James M. Rehg, and Pinar Yanardag
06Hot off the press
What are we cooking up?
Fresh work from GEMLAB, check out our latest projects below.
AdaState: Self-Evolving Anchors for Streaming Video Generation
VideoMLA: Low-Rank Latent KV Cache for Minute-Scale Autoregressive Video Diffusion
Aligning Latent Geometry for Spherical Flow Matching in Image Generation
From Zero to Hero: Training-Free Custom Concept Spawning in World Models
07The team
Meet us in person
Come say hello — reach any of us by email or on the web.
PY
Pinar Yanardag
Principal Investigator
HY
Hidir Yesiltepe
PhD Student
YD
Yusuf Dalva
PhD Student
TM
Tuna Han Salih Meral
PhD Student
TK
Tahira Kazimi
PhD Student
RA
Ritika Allada
GEMLAB Alumna
Other lab members who are not at CVPR but contribute greatly to all of our research can be found on our team page.
08Collaborate
Collaborate with GEMLAB
GEMLAB collaborates with leading industry research labs on generative AI, controllability, creativity, and responsible media. We're always open to new partnerships — reach out to start a conversation.
Selected research partners
Google Research
Snap Research
Qualcomm Research
Adobe Research
& more
Get in touch