Arman Zarei

I am a fourth-year Computer Science Ph.D. student at the University of Maryland, advised by Soheil Feizi. I am currently a Research Intern at Meta, where I work on autoregressive video diffusion models, and have previously interned at Adobe, Netflix, ByteDance TikTok, and Pinterest.

My research broadly focuses on understanding and interpreting image and video generation and editing models, identifying their limitations, and developing methods that provide users with more reliable and fine-grained control over model outputs. I am particularly interested in controllable image editing, compositional generation, localizing and editing knowledge in text-to-image generative models, and autoregressive video diffusion models.

My past research has also spanned Deep Learning Robustness, 3D Vision, and applications of Machine Learning to neuroscience, including improving seizure detection systems.

Email / Google Scholar / Linkedin / Github / CV

SliderEdit++: Improved Slider Trajectories in Instruction-Based Image Editing

Arman Zarei, Sayan Nag, Bryan Wang, Franck Dernoncourt, Varun Manjunatha, Nedim Lipka, Soheil Feizi, Samyadeep Basu

Under Review

Project, Paper, Code

Our paper introduces SlideBench, the first thorough benchmark with robust metrics for evaluating sliders in instruction-based image editing, and SliderEdit++, which regularizes the intermediate trajectory of slider-based control in instruction-based image editing for smooth, monotonic, and disentangled control, achieving state-of-the-art continuous editability while being 3x to 20x faster than prior slider-based approaches.

AgentComp: From Agentic Reasoning to Compositional Mastery in Text-to-Image Models

Arman Zarei, Jiacheng Pan, Matthew Gwilliam, Soheil Feizi, Zhenheng Yang

Under Review

Project, Paper

Our paper introduces AgentComp, an agentic orchestration framework that autonomously builds compositional training data and trains text-to-image models to better distinguish compositionally similar prompts and images, resulting in stronger and more reliable compositional generation without sacrificing image quality.

Improving Compositional Attribute Binding in Text-to-Image Generative Models via Enhanced Text Embeddings

Arman Zarei*, Keivan Rezai*, Samyadeep Basu, Mehrdad Saberi, Mazda Moayeri, Priyatham Kattakinda, Soheil Feizi

Preprint

Project, Paper

Our paper demonstrates that text-to-image generative models often fail at accurately composing attributes and relationships due to sub-optimal text conditioning by the CLIP text-encoder, and we show that significant compositional improvements can be achieved by fine-tuning a simple linear projection on CLIP's representation space.

PhytoOracle: Scalable, modular phenomics data processing pipelines

Emmanuel M. Gonzalez, Ariyan Zarei, Nathanial Hendler, Travis Simmons, Arman Zarei, Jeffrey Demieville, Robert Strand, Bruno Rozzi, Sebastian Calleja, Holly Ellingson, Michele Cosi, Sean Davey, Dean O. Lavelle, Maria Jose´ Truco, Tyson L. Swetnam, Nirav Merchant, Richard W. Michelmore, Eric Lyons, Duke Pauli

Frontiers in Plant Science 2023

Paper

PhytoOracle (PO) introduces modular, scalable pipelines for processing large volumes of phenomics data, improving efficiency, enabling data fusion, and supporting multi-system trait extraction, with broad applicability across species and various data sources, including drone data.