Mark Yatskar

Mark Yatskar is an Assistant Professor in the Department of Computer and Information Science at the University of Pennsylvania. His research focuses on the intersection of natural language processing, computer vision, and fairness in machine learning. He earned his PhD from the University of Washington under advisors Luke Zettlemoyer and Ali Farhadi, and previously worked as a Young Investigator at the Allen Institute for Artificial Intelligence.

Education:

• PhD in Computer Science, University of Washington (Advisor: Luke Zettlemoyer & Ali Farhadi)

Research Interests: Yatskar's work explores how language can structure visual perception and mitigate human biases in machine learning systems. Key themes include:

• Natural language as a scaffold for visual intelligence
• Bias characterization and control in machine learning systems

His lab currently investigates projects like language-guided bottlenecks, annotator cognitive heuristics, and gender bias amplification.

Teaching:

• CIS 5300: Computational Linguistics (2021-2024)
• CIS 7000: Language and Vision (2020)
• CIS 6300: Efficient NLP (2023, 2025)

Awards:

• Best Paper Award at EMNLP (Gender Bias Amplification Research)

Advising & Grants: Yatskar advises a team of PhD/Master's students and actively seeks motivated researchers. His group has explored funding in areas like interpretable AI, multimodal reasoning, and dataset bias mitigation.

Labs/Teams: Leads the Penn NLP & Vision Lab, focusing on projects like MolMo/PixMo open models, ViUniT visual unit tests, and bias mitigation frameworks.