Join us to share your thoughts about teaching, learning, and AI!

The landscape of higher education is rapidly evolving with the integration of Artificial Intelligence (AI). Through the Institute on AI, Pedagogy, and the Curriculum with AAC&U, we are exploring ways that we can better address AI in teaching and learning. We want to hear your experiences, your concerns, and your ideas.

This is an open discussion for all faculty and staff to share their perspectives on the opportunities and challenges AI presents in our academic environment.

We'll be exploring critical questions like:
  • In the age of AI, what are the opportunities you see for enriching the classroom and curriculum? How can it enhance student learning or your professional practice?
  • What are the most significant challenges and concerns that AI raises for you regarding academics, student integrity, or your workload?
  • What resources (tools, training, technical support, policy guidance, etc.) do you need to feel confident and successful in the age of AI?
Date: Monday, December 1st
Time: 12:30pm-1:45pm
Location: West Campus - Melville Library, Special Collections Seminar Room (the room is to the left at the top of the first flight of stairs from the Melville lobby)
or
Date: Wednesday, December 3rd
Time: 10:30am-11:45am
Location: East Campus - HSC 2-154

Please register in advance so we can confirm the room.

Note: Videos will not be shared publicly and comments will only be shared in aggregate.

Your input is vital. From pedagogy to assessment, your insights will be critical. We look forward to a thoughtful and productive conversation!
  • Dr. Rose Tirotta-Esposito (Assistant Provost; Director of CELT)
  • Dr. Elizabeth Hewitt (Associate Professor in the Department of Technology and Society (DTS) in the College of Engineering and Applied Sciences)
  • Chris Kretz (Associate Librarian and Head of Academic Engagement at SBU Libraries)
  • Prof. Rajiv Lajmi (Assistant Professor in the School of Health Professions and Chair of Applied Health Informatics)
  • Dr. Matthew Salzano (Assistant Professor in the Department of Communication in the School of Communication and Journalism)
Abstract: Many unresolved legal questions over LLMs and copyright center on memorization: whether specific training data have been encoded in the model's weights during training, and whether those memorized data can be extracted in the model's outputs. While many believe that LLMs do not memorize much of their training data, recent work shows that substantial amounts of copyrighted text can be extracted from open-weight models. However, it remains an open question if similar extraction is feasible for production LLMs, given the safety measures these systems implement. We investigate this question using a two-phase procedure: (1) an initial probe to test for extraction feasibility, which sometimes uses a Best-of-N (BoN) jailbreak, followed by (2) iterative continuation prompts to attempt to extract the book. We evaluate our procedure on four production LLMs -- Claude 3.7 Sonnet, GPT-4.1, Gemini 2.5 Pro, and Grok 3 -- and we measure extraction success with a score computed from a block-based approximation of longest common substring (nv-recall). With different per-LLM experimental configurations, we were able to extract varying amounts of text. For the Phase 1 probe, it was unnecessary to jailbreak Gemini 2.5 Pro and Grok 3 to extract text (e.g, nv-recall of 76.8% and 70.3%, respectively, for Harry Potter and the Sorcerer's Stone), while it was necessary for Claude 3.7 Sonnet and GPT-4.1. In some cases, jailbroken Claude 3.7 Sonnet outputs entire books near-verbatim (e.g., nv-recall=95.8%). GPT-4.1 requires significantly more BoN attempts (e.g., 20X), and eventually refuses to continue (e.g., nv-recall=4.0%). Taken together, our work highlights that, even with model- and system-level safeguards, extraction of (in-copyright) training data remains a risk for production LLMs.

Speaker: Xinyue

Location: CS2311
Prof. Eugene A. Feinberg, from the Department of Applied Mathematics and Statistics, presents, Recent Developments in Markov Decision Processes Relevant to AI on April 4 at 4p. The talk discusses recent developments in Markov Decision Processes potentially relevant to artificial intelligence. These developments include complexity estimations for exact and approximate algorithms, decision making with incomplete information and multiple criteria, and continuity properties of optimal values and expectations. Dr. Eugene A. Feinberg is currently Distinguished Professor in the Department of Applied Mathematics and Statistics at Stony Brook University. He is an expert on applied probability, stochastic models of operations research, Markov decision processes, and on industrial applications of operations research and statistics. He has published more than 150 papers and edited the Handbook of Markov Decision Processes. His research has been supported by NSF, DOE, DOD, NYSTAR (New York State Office of Science, Technology, and Academic Research), NYSERDA (New York State Energy Research and Development Authority) and by industry. He is a Fellow of INFORMS (The Institute for Operations Research and Management Sciences) and has received several awards including 2012 IEEE Charles Hirsh Award for developing and implementing smart grid technologies, 2012 IBM Faculty Award, and 2000 Industrial Associates Award from Northrop Grumman. Dr. Feinberg is an Associate Editor for Mathematics of Operations Research and for Applied Mathematics Letters. He is an Area Editor for Operations Research Letters. Refreshments will be provided
Join the Department of Computer Science as we welcome Lyle Ungar, University of Pennsylvania, who will be delivering a lecture on 'Measuring Cultural Variation using Natural Language Processing.' When: 11/08/24 @ 2:30 PM Where: New Computer Science Building, Room 120. Reception to follow. Abstract: Cultures vary widely in how they view the world, for example being more individualist or collectivist. Such cultural differences are, of course, reflected in the words that people use. We first show a variety of ways in which multilingual language models are not multicultural; they speak Hindi or Mandarin, but still think like Americans. In contrast, we then present a scalable method that uses embedding-derived lexica to successfully measure regional variation in culture. Bio: Lyle Ungar is a Professor of Computer and Information Science at the University of Pennsylvania, where he also holds secondary appointments in Psychology, Bioengineering, Genomics and Computational Biology, and Operations, Information and Decisions. His group uses natural language processing and explainable AI for psychological research, including analyzing social media and cell phone sensor data to better understand the drivers of physical and mental well-being. They are currently building socio-emotionally sensitive GPT-based tutors and coaches.
Abstract: Jailbreak attacks circumvent LLMs' built-in safeguards by concealing harmful queries within adversarial prompts. While most existing defenses attempt to mitigate the effects of adversarial prompts, they often prove inadequate as adversarial prompts can take arbitrary, adaptive forms. This paper introduces RobustKV, a novel jailbreak defense that takes a fundamentally different approach by selectively removing critical tokens of harmful queries from key-value (KV) caches. Intuitively, for an adversarial prompt to be effective, its tokens must achieve sufficient `importance' (measured by attention scores), which consequently lowers the importance of tokens in the concealed harmful query. Therefore, by carefully evicting the KVs of low-ranked tokens, RobustKV minimizes the harmful query's presence in the KV cache, thus preventing the LLM from generating informative responses. Extensive evaluation using benchmark datasets and models demonstrates that RobustKV effectively counters state-of-the-art jailbreak attacks while maintaining the LLM's performance on benign queries. Notably, RobustKV creates an interesting effectiveness-evasiveness dilemma for the adversary, leading to its robustness against adaptive attacks.

Speaker: Tanqiu Jiang

Where: NCS 220 and Zoom (https://stonybrook.zoom.us/j/6406956411)

Abstract: Pre-trained diffusion and flow matching models have made visual generation remarkably powerful, enabling high-fidelity synthesis of images and videos from natural language prompts. However, their behavior is still largely dictated by the pre-training data distribution and likelihood objective, which do not directly encode downstream desiderata such as fine-grained semantic alignment, controllability, or realism. This gap motivates post-training: starting from a base generator and further optimizing it with additional supervision signals derived from human or reward model preferences.This work presents post-training for visual generative models through two complementary case studies. First, Hummingbird addresses the problem of fine-grained contextual alignment in image-text-to-image generation. We introduce a multimodal context evaluator that scores the consistency between rich contextual descriptions and generated images, capturing fine-grained alignment beyond global CLIP similarity. By directly backpropagating these differentiable rewards through the diffusion sampler, Hummingbird substantially improves semantic faithfulness while preserving high visual quality.
Second, PISCES tackles post-training for text-to-video generation, where alignment is inherently semantic-spatio-temporal. We show that naive VLM-based rewards suffer from distributional mismatch and token-level misalignment, leading to reward hacking and suboptimal optimization. PISCES introduces a bi-objective, Optimal Transport (OT)-aligned reward module: distributional OT using Neural Optimal Transport to align text and video embedding distributions, and discrete, partial OT over a spatio-temporal cost matrix to capture semantic alignment at the token level. These rewards are integrated into both direct backpropagation and GRPO-style optimization to post-train state-of-the-art text-to-video generators. Together, Hummingbird and PISCES provide a unified view of how carefully designed visual reward models, coupled with OT-based representation alignment, can reliably improve the downstream behavior of pre-trained image and video generators.

Speaker: Minh Quan Le

Location: NCS 220

Zoom: https://stonybrook.zoom.us/j/94798224254?pwd=CFraer25qnpORbJ14aAVHRwaSJOjJM.1
Are you concerned about AI issues with your asynchronous online courses? Is your fully online course vulnerable to AI plagiarism? Do you want to engage your online students using AI? Discover the future of education with our AI-powered solutions designed specifically for online asynchronous courses. This innovative approach uses artificial intelligence to transform the way courses are delivered, making learning more personalized, engaging, and effective.

Register here: https://stonybrook.zoom.us/meeting/register/RD94cHiHRwCj6xNkCZqNEg

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
AI can help you write, you hear. AI can save you time, leverage your skills, enhance your productivity. . . . But you also hear: AI output is not reliable, not adequate for advanced tasks/learning, not ethical to use -- you could get in deep trouble for using AI tools without adequate mastery and caution. Which way is it?
Come join this hands-on workshop where you will explore AI tools and their affordances. Engage in writing tasks to learn how to use AI tools effectively and responsibly.
Sign up for a seat now: https://docs.google.com/forms/d/e/1FAIpQLSd0iDTKkTYnkxFd4LkgqbtP97zQSS4FI_MiPVm7p6IY5SGwSg/viewform