AI Innovation Institute

Abstract: Computer vision seeks to extract semantic and geometric information from images and videos, serving as the perceptual foundation for intelligent systems such as robots and autonomous vehicles. Over the past decade, deep learning has driven remarkable progress in the field, advancing capabilities from 2D recognition to 3D reconstruction. However, the current purely data-driven paradigm faces fundamental challenges, including data inefficiency, curse of high dimensionality, and limited understanding of visual entities beyond individual objects.

In this talk, I will present my recent research on modeling and learning rich visual structures to address these challenges. First, I will introduce a novel framework that integrates explicit visual dependency modeling with deep learning for 2D and 3D dense prediction. Next, I will demonstrate how unfolding the manifold structure of visual data enables unsupervised semantic segmentation. Finally, I will present a recent project that represents, parses, and learns the geometric compositionality of 3D objects to facilitate self-supervised part-whole reconstruction. Through these efforts, I aim to bridge the gap between data-driven deep learning and visual structure modeling, paving the way for more efficient, generalizable, and interpretable computer vision models.

Bio: Dr. Wei Tang is an Assistant Professor in the Department of Computer Science at the University of Illinois Chicago (UIC). He obtained his Ph.D. in Electrical Engineering from Northwestern University, where his dissertation was honored with a Best Dissertation Award. His research interests include computer vision, digital image processing, and machine learning. Dr. Tang has served as an associate editor for several international journals, including Pattern Recognition and Machine Vision and Applications, and as an area chair for leading conferences, including CVPR, ICCV, and WACV. His research has been funded by the National Science Foundation (NSF) and industry partners such as Motorola and Wormpex AI Research.


Location: NCS 115

Zoom: https://stonybrook.zoom.us/j/4624091659?omn=95178138684&jst=3

Fall 2025, Mondays 2 to 3:20 pm, NCS 220 and Zoom link to be announced soon.

The seminar will be jointly taught by Prof. Dimitris Samaras samaras@cs.stonybrook.edu.

The overall purpose of this seminar is to bring together people with interests in Computer Vision theory and techniques and to examine current research issues. This course will be appropriate for people who already took a Computer Vision graduate course or already had research experience in Computer Vision.

To enroll in this course, you must either: (1) be in the Ph.D. program or (2) receive permission from the instructors.

Each seminar will consist of multiple short talks (around 15 minutes) by multiple students. Students can register for 1 credit for CSE656. Registered students must attend and present a minimum of 2 talks. Registered students must attend in person. Up to 3 absences will be excused. Everyone else is welcome to attend.

Understand how AI can be used for your marketing.

Register here.

A talk by Jerome Zhengrong Liang entitled, Machine Learning from Original Images to Texture Patterns: A Paradigm Shift from Non-Medical Application to Medical Diagnosis. Abstract: Artificial intelligence (AI) research for medical diagnosis started soon after human began to use computer, initially called artificial neural network (ANN) and now convolutional neural network (CNN). ANN has been mainly explored to classify the experts' handcrafted features from the original (or raw) images, while CNN has been mainly explored directly on the raw images for both tasks of extracting abstract features and classifying the features. Experimental evidences have been shown that CNN can be trained by a large number of the raw images with experts' scores (or labels) to match or even surpass the experts' performance for both non-medical and medical diagnosis applications. However, the performances of the CNN models as well as the experts on medical diagnosis dropped dramatically when the labels of the raw images were replaced by the corresponding medical pathological reports. Accumulated medical knowledge reveals that the lesion heterogeneity is a footprint of lesion evolution and ecology, and the heterogeneity is an indicator of lesion progress and response to medical intervention. The heterogeneity can be reflected by the image contrast distribution (or texture patterns) across the lesion volume. Image textures have been shown as an effective descriptor of the lesion heterogeneity for computer-aided diagnosis. Can we map the raw images into texture patterns (or images) and train CNN to learn from the texture images? This question is the central theme of this presentation with application to CT Colonography or virtual colonoscopy, a game from AlphaGo to PolypGo. Bio: Jerome Zhengrong Liang, PhD, IEEE Fellow Imaging Research and Informatics Laboratory Department of Radiology, Stony Brook University

CSE 600 Seminar Series | Fall 2025


Abstract: Large reasoning models have demonstrated capabilities to solve competition-level math problems, answer deep research questions, and address complex coding needs. Much of this progress has been enabled by scaling of data: pre-training data to learn vast knowledge, fine-tuning data to learn natural language reasoning, and RL environments to refine that reasoning. In this talk, I will describe the current LLM reasoning paradigm, its boundaries, and the future of LLM reasoning beyond scaling. First, I will describe the state of reasoning models and where I think scaling can lead to some additional (though perhaps limited) successes. I will then shift to discussing more fundamental issues with models that scale will not resolve in the next few years. I will touch on four current limitations: outdated knowledge, generator-validator gaps, limited creativity, and poor compositional generalization. In all cases, fundamental limitations of LLMs or of supervised learning in general make these problems challenging, inviting future study and novel solutions beyond scaling.

Bio: Greg Durrett is an associate professor in the Department of Computer Science and the Center for Data Science at New York University. His research is broadly in the areas of natural language processing and machine learning. Currently, his group's focus is on reasoning about knowledge in text, verifying correctness of generation methods, and studying how to make progress on problems that defy LLM scaling. He is a 2023 Sloan Research Fellow and a recipient of a 2022 NSF CAREER award. He has served in numerous roles for ACL conferences, recently as a member of the NAACL Board since 2024 and as Senior Area Chair for ACL 2025 and EMNLP 2025. He received his BS in Computer Science and Mathematics from MIT and his PhD in Computer Science from UC Berkeley, where he was advised by Dan Klein.

Language shared online through social media or messaging reflects people's thoughts and emotions. Processing this data with Natural Language Processing (NLP) and machine learning can reveal mental health and psychological traits. For example, analyzing Facebook posts enables me to predict depression before it is clinically diagnosed and highlight particular symptoms. At the population level, billions of geo-tagged Tweets can be used to monitor health risk patterns, including depression and anxiety trends across communities. Beyond assessment, I'm using Large Language Models (LLMs) to improve mental health care, including training therapists and assisting with Cognitive Behavioral Therapy. These applications of NLP and Al may lead to earlier and more effective interventions and improved access for underserved populations. Speaker: Johannes Eichstaedt, Ph.D. Assistant Professor, Psychology & Human-Centered Al, Stanford University

Abstract: Modern language agents often need to solve tasks requiring long-horizon, multi-turn interactions, where they retrieve external information, adapt to observations, and answer interdependent queries. Yet, most LLM systems rely on full-context prompting, appending all past turns regardless of their relevance. This leads to un-bounded memory growth, increased computational costs, and degraded reasoning performance on out-of-distribution input lengths due to LLM forgetting the context. We introduce MEM1, an end-to-end reinforcement learning framework that enables agents to operate with constant context size when solving long multi-turn tasks. At each turn, MEM1 updates a compact shared internal state that jointly supports memory consolidation and reasoning. Leveraging reinforcement learning (RL) and rollout trajectory truncation, we train a MEM1 agent to develop internal states that integrate prior memory with new observations from the environment while strategically discarding irrelevant or redundant information. Experiments across three domains, including internal retrieval QA, open-domain web QA, and multi-turn web shopping, show that MEM1-7B improves performance by 3.5x while reducing memory usage by 3.7x compared to Qwen2.5-14B-Instruct on an augmented multi-hop QA dataset with 16 objectives in each task, and generalizes beyond the training horizon. Our results demonstrate the promise of reasoning-driven memory consolidation as a scalable alternative to existing solutions for training long-horizon task-solving agents that involve multiple interactions, where both efficiency and performance are optimized.

Speaker: Yiyang Feng

Location: CS2311

AI3, SBU Libraries and IACS present
at International Love Data Week
sponsored by The Office of the Provost and
Educational and Institutional Effectiveness (EIE)

Special Talk and Panel Discussion

How I Learned to Stop Worrying and Love AI (For Now)

with Paul Fain from The Job and Work Shift

A reporter's take on what we know--and what we don't know--about AI's emerging impacts on the labor market. The discussion will include the latest research from economists and the AI labs themselves about how workers are using AI, and current thinking among experts on how the tech's rapid deployment will play out across job roles, industries, and regions.

Panel discussion to follow with:

• Lav Varshney, Della Pietra Infinity Professor and inaugural director of the AI Innovation Institute
• Nicholas Johnson, Director of AI, SBU Libraries
• Marianna Savoca, Associate Vice President for Career Readiness and Experiential Education

Paul Fain is co-founder of Work Shift, editor of the must-read newsletter, The Job, and host of The Cusp podcast. A veteran higher education reporter, Paul is perhaps the nation's top journalist focused on connections between education and work. He started Work Shift after a decade as a senior reporter and then news editor at Inside Higher Ed, where he led the outlet's coverage of low-income and first-generation students, college completion, community colleges, federal policy, and emerging models of higher education. He also was the founding host of the successful podcast, The Key with Inside Higher Ed, and has contributed chapters for books on innovation in higher education, published by the Harvard University Press and the Stanford University Press. Earlier in his career, Paul was a senior reporter at The Chronicle of Higher Education.

Limited Seats!

Registration is required.

Join Zoom Meeting https://stonybrook.zoom.us/j/511954566?pwd=d3hGdHlGZVJISnV3d2pRcnBRbFBJZz09 (ID: 511954566, password: fxnmj4) Join by phone (US) +1 646-876-9923 Join using SIP 511954566@zoomcrc.com (passcode: 542040) Joining instructions: https://www.google.com/url?q=https://applications.zoom.us/addon/invitation/detail?meetingUuid%3DDBgsLRhCTQCGwD9DAxf7mw%253D%253D%26signature%3D02e2f3393481f8bc18cbb5520e174a6b8ceee7e9e58c9dca9d0aa4db9a63279a&sa=D&usg=AOvVaw1e3Mi7F_8TdXl5XstjsrIx Password: fxnmj4