Abstract: Large language models are prone to memorizing some of their training data. Memorized (and possibly sensitive) samples can then be extracted at generation time by adversarial or benign users. There is hope that model alignment---a standard training process that tunes a model to harmlessly follow user instructions---would mitigate the risk of extraction. However, we develop two novel attacks that undo a language model's alignment and recover thousands of training examples from popular proprietary aligned models such as OpenAI's ChatGPT. Our work highlights the limitations of existing safeguards to prevent training data leakage in production language models.

Speaker: Pegah Alipoormolabashi

Location: CS2311

You are cordially invited to attend the biweekly Brookhaven AI Mixer (BAM). BAM includes one short talk on AI research happening at BNL, followed by an open mixer over coffee and snacks for everyone to network and discuss all things AI. The first half hour will consist of presentations that will be available via ZOOM, and the second half hour will be for in person only networking.

Abstract: Two-dimensional (2D) materials such as graphene, hBN, and TMDs offer atomically sharp interfaces and unprecedented tunability when vertically assembled into van der Waals heterostructures. These stacks have enabled discoveries ranging from moiré superconductivity and correlated insulators to quantum emitters and next-generation nanoelectronic devices. Yet constructing high-quality heterostructures remains largely artisanal: researchers manually identify exfoliated flakes, align a polymer stamp by eye, and finely adjust temperature and contact geometry through tacit skill. This manual workflow is difficult to reproduce, scales poorly, and prevents systematic exploration of the enormous combinatorial space of materials, twist angles, and interfacial conditions. AutoLab is an autonomous platform that translates this tacit human expertise into programmable, feedback-driven control. Instead of pressing flakes with predefined trajectories, AutoLab uses machine vision to detect polymer-wafer contact, dynamically regulates contact evolution through closed-loop actuation and temperature control, and captures high-quality flakes with the cleanliness and precision of expert manual fabrication. The system integrates perception, decision making, and motion planning into a single robotic framework, enabling reproducible stacking, wafer-level coverage, and accelerated discovery. Beyond 2D materials, AutoLab illustrates a broader paradigm for AI-native scientific automation: codifying human experimental reasoning into algorithms that interrogate data in real time, adaptively adjust instrumentation, and generate scalable, high-fidelity datasets. Such platforms could generalize to diverse research domains--quantum device fabrication, optical alignment, surface science, autonomous microscopy, and other workflows where expert intuition currently limits throughput and reproducibility. By bridging artisanal manipulation and robotic autonomy, AutoLab points toward a future where scientific discovery is accelerated by machines that not only execute instructions, but learn, respond, and collaborate with human scientists.

Biography: Dr. Yutao Li is a research associate from Department of Condensed Matter Physics and Material Science, Brookhaven National Laboratory. He has 8 years of experience in 2D material sample fabrication, and investigation in their electronic transport, optical and mechanical properties.

Join us every other Tuesday at noon in CDSD's Training Room (building 725, 2nd floor) to learn about interesting AI methods and applications, engage with potential collaborators, prepare for pending FASST funding calls, and build a community of AI for Science at BNL.

Location: CDS, Bldg. 725, Training Room

Join ZoomGov Meeting: https://bnl.zoomgov.com/j/1604383624?pwd=ffQ5cUPNxTI7nzClKQO6cnsNbhF9Vf.1

Meeting ID: 160 438 3624
Passcode: 558449

This is Stony Brook's quantum moment. Join us for a spotlight on the core achievements and research excellence of faculty across the Colleges of Arts and Sciences (CAS), and Engineering and Applied Sciences (CEAS) - and their collaborative advancements in quantum science and technology. Learn about the real world impact of their enduring work, their leadership in translating foundational science into entrepreneurial opportunities, and their impetus for making connections to next generation innovation.

Presented by: Catherine Chen, Ph.D., Research Development Associate

Welcome remarks: President Andrea Goldsmith

Panel moderators: Dean David Wrobel, CAS, and Dean Andrew Singer, CEAS

Presentations and panel featuring our faculty:

  • Jennifer Cano, CAS, Physics and Astronomy

  • P. Scott Carney, CEAS, Mechanical Engineering

  • Hyeongrak Chuck Choi, CEAS, Electrical and Computer Engineering

  • Eden Figueroa, CAS, Physics and Astronomy

  • Humanshu Gupta, CEAS, Computer Science

  • Angela Kelly, CAS, Physics and Astronomy

Location: Theatre at the Charles B. Wang Center, Stony Brook University

Reserve your tickets by March 26!

Abstract: Humans perceive the world around them by recognizing global patterns and structures such as object parts, branches, their spatial arrangement, and so on. Most deep learning models, however, take a fundamentally local approach. They process images pixel-by-pixel rather than focusing on structures as a whole. While these models indeed perform well on many tasks, the local (pixel-level) versus global (structure-level) disconnect makes them harder to interpret and control.

Topology, in a general sense, is a mathematical language for describing structure. It delineates how different parts of an image relate to one another, capturing both individual structures and their overall layout. Preserving topology enforces structural correctness and, by extension, semantic validity.

In this thesis, we investigate how topological constraints can be used to bridge the gap between local and global understanding. We use topology to inform the design of deep learning models that are explicitly structure-aware. Our thesis focuses on dense prediction tasks, which include image segmentation, uncertainty estimation, and generative modeling. First, we introduce a topological interaction module for semantic segmentation that encodes containment and exclusion constraints directly into the learning process. This preserves anatomical hierarchies and improves multi-class consistency. Next, since segmentation models can never be truly perfect, we address the need for reliable uncertainty estimation to identify error-prone regions. Unlike conventional pixel-wise uncertainty maps, which tend to be noisy and difficult to interpret, we propose reasoning at the level of structural units--branches and connections--which are more visually discernible and actionable. Finally, we leverage topology for generative modeling. We propose a topology-guided diffusion framework that can be controlled using structural attributes like object count and connectivity.

Together, these contributions establish a unified approach to topology-informed, structure-preserving dense prediction models. By integrating topological reasoning with deep networks, this thesis advances models that are not only accurate, but also structurally consistent, interpretable, and controllable. The results from this thesis have been published in ECCV, NeurIPS, and ICLR.

Speaker: Saumya Gupta

Location: New Computer Science (NCS) 120


Zoom: https://stonybrook.zoom.us/j/93643318604?pwd=kv8DagpbayzizivU29UCYItnlzlYRM.1&jst=2
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)
The Antonija Prelec Memorial Committee in collaboration with Stony Brook University Libraries are very excited to bring you the 2019 Prelec Memorial Lecture! This year, we are pleased to announce our speaker is Patricia Flatley Brennan, RN, PhD, Director of the National Library of Medicine.

No registration required. Find more information here.