Yusuf Dalva

  About Me

I am a Ph.D. student at Virginia Tech, Blacksburg, where I am affiliated with Sanghani Center for Artificial Intelligence and Data Analytics. I am working on making generative models more controllable and customizable to further enhance control over them. Before joining Virginia Tech, I obtained my M.Sc. & B.Sc. degree from Deprtament of Computer Engineering at Bilkent University (M.Sc. Thesis).

At Virginia Tech, I am currently working on the controllability aspect of generative models under the supervison of Pinar Yanardag. In the past, I have been fortunate to work with Aysegul Dundar at Bilkent University, and Yijun Li at Adobe Research.

  Research Interests

• Text-to-image generation and editing
• Customization of generative modesls
• Representations learned by generative models

  Recent Research

LoRAShop: Training-Free Multi-Concept Image Generation and Editing with Rectified Flow Transformers

Yusuf Dalva, Hidir Yesiltepe, Pinar Yanardag

NeurIPS 2025 (Spotlight - top 3%)

Paper, Project Page

FluxSpace: Disentangled Semantic Editing in Rectified Flow Transformers

Yusuf Dalva, Kavana Venkatesh, Pinar Yanardag

CVPR 2025

Paper, Project Page

NoiseCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions in Diffusion Models

Yusuf Dalva, Pinar Yanardag

CVPR 2024 (Oral - top 0.8%)

Paper, Project Page

For the full list of publications, visit the Publications Page and my Google Scholar profile.