Meta Pre-Training with Augmentations to Generalize Neural Network Processing for Domain Adaptation
U.S. Patent, 2025 · Developed with Qualcomm AI Research
Saeed Vahidian, M. Bhat, D. Das, S. Steve Han, and F. Murat Porikli
Generative and Multimodal Intelligence
Saeed Vahidian
Assistant Professor, School of Electrical and Computer Engineering, Oklahoma State University
I am currently a tenure-track assistant professor in the School of Electrical and Computer Engineering at Oklahoma State University. Before joining OSU, I completed my postdoctoral research at Duke University, where I worked with Prof. Yiran Chen, Director of the NSF AI Institute for Edge Computing (Athena).
I received my Ph.D. in Electrical and Computer Engineering from the University of California San Diego under the supervision of Prof. Bill Lin.
My research lies at the intersection of multimodal learning, large language models (LLMs), generative AI, agentic AI, synthetic data, and robust learning. I develop reliable and efficient learning systems that connect foundational AI research with real-world applications.
Openings: I am recruiting two fully funded Ph.D. students to work on multimodal learning, LLMs, generative AI, and agentic AI. If you are interested, please email me at saeed.vahidian@okstate.edu with your CV and academic transcripts.
Openings: Two fully funded Ph.D. positions are available in multimodal learning, large language models (LLMs), generative AI, and agentic AI. Learn more.
News
- 08/2026Joined Oklahoma State University as an Assistant Professor in Electrical and Computer Engineering.
- 08/2026Recruiting two fully funded Ph.D. students in multimodal learning, LLMs, generative AI, and agentic AI.
- 05/2026CoreInfer was accepted at ICML 2026.
- 01/2025Our work on group distributionally robust dataset distillation was accepted at ICLR 2025.
- 06/2024Two papers were accepted at CVPR 2024.
- 06/2024Organized and chaired the first Workshop on Dataset Distillation for Computer Vision at CVPR 2024.
Research Vision
Multimodal Learning and Vision–Language Models
Multimodal foundation models that generate and learn from text, images, video, and audio, with an emphasis on interpretable reasoning and compute-efficient deployment.
Large Language Models (LLMs) and Generative AI
Developing efficient and reliable foundation models, controllable generative systems, multimodal synthetic-data pipelines, and methods that improve how large models learn and adapt.
Agentic AI and Autonomous Intelligent Systems
Designing AI agents that reason, plan, use tools, collaborate, and learn from feedback while remaining dependable, interpretable, and aligned with their intended objectives.
Robust, Efficient, and Data-Centric Learning
Studying robust optimization, dataset distillation, synthetic data, federated learning, and efficient inference under limited data, compute, communication, and privacy budgets.
Innovation & Intellectual Property
Honor, Leadership & Service
- Invitation from NASA to collaborate on a distributed learning project (2022)
- Primary Organizer and Chair, first CVPR Workshop on Dataset Distillation for Computer Vision (2024)
- Co-Organizer and Chair, third CVPR Workshop on Federated Learning for Computer Vision (2024)
- Committee Member, second CVPR Workshop on Federated Learning for Computer Vision (2023)
- Reviewer, CVPR, ICML, NeurIPS, and other leading venues
Publications
Dataset Distillation from First Principles: Integrating Core Information Extraction and Purposeful Learning
Journal of Machine Learning Research, 2025
Efficient Distribution Similarity Identification in Clustered Federated Learning via Principal Angles
AAAI 2023
NeurIPSRethinking Data Heterogeneity in Federated Learning: Introducing a New Notion and Standard Benchmarks
NeurIPS 2022
FLIS: Clustered Federated Learning via Inference Similarity for Non-IID Data Distribution
IEEE Open Journal of the Computer Society, 2023
Personalized Federated Learning by Structured and Unstructured Pruning under Data Heterogeneity
IEEE ICDCS Workshops, 2021
Select to Better Learn: Fast and Accurate Deep Learning Using Data Selection from Nonlinear Manifolds
CVPR 2020
Invited Talks
- Toward Explainable and Ethical Generative AI — University of Houston, Department of Computer Science (2026)
- Synthetic Data Generation and Resilient Learning Structure: Foundations for Scalable, Robust, and Eco-Aware AI Systems — University of Texas at Dallas, Department of Computer Science (2025)
- Synthetic Data Generation and Resilient Learning Structure: Foundations for Scalable, Robust, and Eco-Aware AI Systems — University of Arizona, Department of Electrical and Computer Engineering (2025)
- Sparse Convex Optimization Methods for Machine Learning and Deep Learning — University of California San Diego, Department of Mathematics (2020)
Teaching
AI Systems
Electrical and Computer Engineering, Oklahoma State University — Fall 2026.
Guest Lecturer, ECE 284: Special Topics in Computer Engineering
University of California San Diego — Spring 2022. Instructor: Prof. Bill Lin.
Students & Research Mentoring
Duke University
- Qinsi WangPh.D. studentCoauthored publication at ICML 2026
- Yuqi JiaPh.D. studentCoauthored publication at ECCV 2024
- Hinrik GudmundssonMaster’s studentSummer research intern, 2023
- Steven GuoMaster’s studentSummer research intern, 2023
- Tianlong ChenUndergraduate studentSummer research intern, 2023
- Yuandong ZhangUndergraduate studentSummer research intern, 2023
University of California San Diego
- Weijia WangPh.D. studentCoauthored publications at CVPR, ICCV, and AAAI
- Sreevatsank KadaveruMaster’s studentCoauthored publication at ICCV 2023
- Woonjoon BaekMaster’s studentCoauthored publication at ICCV 2023
Prospective Students
Ph.D. Openings
Two fully funded Ph.D. positions at Oklahoma State University
I am seeking highly motivated students interested in:
- Multimodal learning and vision–language models
- Large language models (LLMs) and generative AI
- Agentic AI and autonomous intelligent systems
- Robust, efficient, and data-centric learning
Interested in joining the group?
Prospective students should have a strong background or demonstrated research interest in machine learning, deep learning, computer vision, natural language processing, or a related field. Strong programming and mathematical skills are expected; experience with Python and PyTorch is highly desirable.
If you are interested in working with my group, please send me a short introductory email with your CV and academic transcripts. You may also include links to relevant publications or projects.
Contact me about joining the group