Aysan Aghazadeh

I’m a PhD candidate in Computer Science at the University of Pittsburgh, working under the supervision of Dr. Adriana Kovashka and a Computer Vision intern at Apple - Camera ML.

My research interests are Computer Vision, Multimodal Reasoning, and Generative AI. My current research is on:

  • Generation and Evaluation of Persuasive Advertisement Images
    • Persuasiveness, Creativity, and alignment evaluation
    • Image generation for abstract text inputs
  • Understanding Advertisement Images
    • Complex and Multi-step reasoning on images
    • Atypicality understanding of the images

News

  • [April 2026] Our workshop, VisPer: Visual Persuasion has been accepted to ECCV 2026.
  • [February 2026] I am joining Apple as a Computer Vision Intern in March 2026.
  • [August 2025] Our paper Face of Persuasion: Analyzing Bias and Generating Culture-Aware Ads is accepted to EMNLP 2025.
  • [June 2025] Our paper CAP: Evaluation of Persuasive and Creative Image Generation (link) is accepted to ICCV 2025.
  • [June 2025] I passed my proposal on topic of Understanding, Evaluating, and Generating Advertisement Content.
  • [May 2025] I am joining Google as a Student Researcher in Fall 2025.
  • [January 2025] I passed my PhD Comprehensive Exam.
  • [December 2024] New preprint. CAP: evaluation of Persuasive and Creative Image Generation (link).
  • [October 2024] Our paper Benchmarking VLMs’ Reasoning About Persuasive Atypical Images (link) is accepted to WACV 2025.
  • [September 2024] New preprint Benchmarking VLMs’ Reasoning About Persuasive Atypical Images (link).
  • [May 2022] Joined Cellanome as a Computer Vision Intern.