Simons Laufer Mathematical Sciences Institute presents...

In 2023, Tudor Achim co-founded Harmonic with Vlad Tenev to build the world's most advanced reasoning engine. Combining formal verification with informal reasoning, Harmonic's formal reasoning model, Aristotle, achieved gold-medal-equivalent performance on the 2025 International Mathematical Olympiad problems. Aristotle integrates three main components: a Lean proof search system, an informal reasoning system that generates and formalizes lemmas, and a dedicated geometry solver.

Achim is also the Co-Founder and former CTO of Helm.ai. He holds a B.S. in Computer Science from Carnegie Mellon University and was a PhD Candidate in Computer Science at Stanford University.

Register here: https://slmath.us10.list-manage.com/track/click?u=d58ee2e82c69809ff037f56b2&id=f07a675f6f&e=f1b6ba91e6

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 Center of Excellence in Wireless and Information Technology (CEWIT) will host the 16th International Conference on Emerging Technologies for a Smarter World (CEWIT2020) virtually on November 5, 2020. The conference will center on the four major fields which are penetrating our business and personal lives: Machine Learning, Artificial Intelligence, Blockchain and Computational Medicine. For more info visit: https://www.cewit.org/.
The 39th Annual AAAI Conference on Artificial Intelligence will be held from February 25 to March 4, 2025 in Philadelphia, Pennsylvania, USA. More information can be found here.
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

Zoom Link: https://stonybrook.zoom.us/j/98533029054?pwd=5FXO6lWGTJssCADEYkYbA7sjaacPRX.1

Meeting ID: 985 3302 9054

Passcode: 436997

Abstract:

Semantic segmentation, the task of assigning a semantic label to each pixel in an image, is a fundamental problem in the field of Computer Vision. with crucial applications in domains like autonomous driving, drone imagery and medical image analysis. Despite advancements in deep learning architectures, state-of-the-art models still heavily depend on large-scale pixel-level annotations, which are costly and time-consuming to acquire. To address this issue, Semi-Supervised Segmentation (SSS) has emerged as a promising solution, leveraging a small set of labeled images alongside a larger corpus of unlabeled data to reduce the annotation burden. In this proposal, I aim to investigate the challenges of SSS and propose approaches to address them. Existing SSS methods rely on a teacher-student framework to generate pseudo-labels for unlabeled images, which are then used for model training. However, this approach presents two major challenges. Pixel-level consistency fails to effectively capture contextual information, and pseudo-labels are noisy, especially in the early stages of training. To address the challenge of noisy pseudo-labels, existing methods rely on confidence-based thresholding to identify reliable pseudo-labels. However, during early training phases, when the model is poorly calibrated, this approach can select high-confidence but noisy pseudo-labels. To address this, we propose a novel approach that reduces reliance on model confidence to select reliable pseudo-labels. Our method employs an ensemble of a segmentation model and an object detection model to select more reliable pseudo-labels, which are then used to weight pseudo-labels using rank statistics, reducing the influence of noisy labels in training. Next, to address both the challenge of capturing contextual information and noisy pseudo-labels I introduce a novel Multi-scale Patch-based Multi-label Classifier (MPMC), which incorporates patch-level contextual information and reduces the impact of noisy pixel pseudo-labels by using the predictions of the patch-level Multi-label classifier to detect noisy labels, enhancing overall segmentation performance. While my work so far has focused on effectively utilizing unlabeled data to improve segmentation performance, as part of our future work, I will explore the use of textual information, such as category descriptions, for segmentation tasks. In limited labeled data scenarios it is more challenging to align visual features with textual features from large language models (LLMs).


Abstract:
Deep learning models have achieved remarkable success across a wide range of computer vision tasks, including image classification, semantic segmentation, etc. However, such success highly relies on a large amount of annotated data, which are expensive to obtain. Moreover, their performance often degrades when there exist distribution shifts between training and test data. Domain Adaptation overcomes these issues by transferring knowledge from a label-rich source domain to a related but different target domain. Despite its popularity, domain adaptation is still a challenging task, especially when the data distribution shifts are severe, while the target domain has no or few labeled data.

In this thesis, I develop four efficient domain adaptation approaches to improve model performance on the target domain. Firstly, inspired by the large-scale pretraining of Vision Transformers, I explore Transformer-based domain adaptation for stronger feature representation and design a safe training mechanism to avoid model collapse in the situation of a large domain gap. Secondly, I observe that source models have low confidences on the target data. To address this, I focus on the penultimate activations of target data and propose an adversarial training strategy to enhance model prediction confidences. Thirdly, I study using weak supervision from prior knowledge about target domain label distribution. A novel Knowledge-guided Unsupervised Domain Adaptation paradigm is devised, and a plug-in module is designed to rectify pseudo labels. Lastly, I step into the task of Active Domain Adaptation, where the labels of a small portion of target data can be inquired. I propose a novel active selection criterion based on the local context and devise a progressive augmentation module to better utilize queried target data. The robustness of domain adaptation approaches, in addition to accuracy, is critical yet under-explored. To conclude the thesis, I empirically study set prediction in domain adaptation using the tool of conformal prediction and conformal training.


Location: New Computer Science Bldg., Room 120
Zoom Link: https://stonybrook.zoom.us/j/92736258273?pwd=ipDdh1CTG6dRYmqa3ltUvooei8OfaT.1Meeting ID: 927 3625 8273
Passcode: 466399

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.

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.

Machine Learning for Seismic Low Frequency Extrapolation

Abstract: The cycle skipping problem that plagues seismic inversion can be mitigated by utilizing low-frequency seismic data, which captures the kinematics of wave propagation, in conjunction with a reasonable initial velocity model. However, seismic sources and receivers are band-limited and cannot provide signals down to 0 Hz. To improve solution of the seismic inverse problem one can synthesize the missing low-frequency content by solving a regression problem using machine learning (ML). The recorded high-frequency (HF) seismic data is the input and the ML models are trained to predict the missing low-frequency (LF) seismic data. Deep learning models utilizing convolutional neural networks (CNNs) and generative adversarial networks (GANs) demonstrate important capabilities for LF extrapolation. However, such models require powerful hardware and careful training. We explore the feasibility of using less costly ML models such as a random forest, Gaussian process surrogates, and gradient boosting as alternatives to computationally expensive deep learning models.

Biography: Sue Minkoff is Chair of Applied Mathematics at Brookhaven National Laboratory. From 2012-2024 she was a Professor of Mathematical Sciences and an Affiliated Professor in the Departments of Sustainable Earth Systems Sciences and Science and Mathematics Education at the University of Texas at Dallas. From 2000-2012 she served on the faculty in the Department of Mathematics and Statistics at the University of Maryland, Baltimore County. She received her doctorate in Computational and Applied Mathematics from Rice University. From 1995-1997 she was a National Science Foundation-Industrial postdoc joint with the University of Texas at Austin and British Petroleum, and from 1997-2000 she held the von Neumann Fellowship in the Mathematics Department at Sandia National Labs. In 2000 Minkoff was promoted to Senior Member of the Technical Staff in Sandia's Geophysics Department. Minkoff's research interests include scientific computing, inverse problems, uncertainty quantification and digital twins modeling, Earth science, and photonics.

Location: CDS, Bldg. 725, Training Room

Join ZoomGov Meeting: https://bnl.zoomgov.com/j/1606848158?pwd=miUtq7OkYL5SNkjbgVb19teZPNennd.1

Meeting ID: 160 684 8158
Passcode: 068399

Abstract: Many scientific and engineering challenges, such as the design of materials or molecules or the control of experimental systems, rely on the existence of fast predictive models that can evaluate potential designs or control policies. Traditionally this has been accomplished through numerical simulation; more recently data-driven machine learning methods have been applied. However, both approaches leave gaps: physical modeling can be accurate and extrapolates well to previously-unstudied conditions, but it is often computationally expensive and relies on physics approximations that may not be valid. Machine learning can generalize from massive amounts of real-world or simulation data, but suffers from physical grounding and extrapolation into new regimes, as well as in settings where large data sets do not exist.
In this talk I explore an intermediate regime, which is hybrid reduced order models: fast simplified physics approximations where some of the unknown or approximated equations are replaced with data-driven machine learning components. Examples include coarse-grained models where the full macroscopic equations cannot be derived from first-principles microscopic equations, multiscale models with unknown closure terms or sub-grid parameterization schemes, and low-order or latent dynamical systems that learn governing equations on a low-dimensional reduced state space. I discuss how such reduced systems can be identified from very limited data, much less than is often needed in traditional machine learning but at much lower time-to-solution than traditional numerical modeling. This facilitates not only system design and control but also uncertainty quantification approaches that search the space of possible equations for predictive models that can explain the data. I will focus on an example from materials science concerning the design of self-assembling block copolymer nanomaterials.

Speaker: Dr. Nathan Urban, Applied Mathematics Department, Brookhaven National Laboratory

Location: Laufer 101

Zoom: https://stonybrook.zoom.us/j/96090260834?pwd=mw8QTHbMOw9oeU9hazZeoq8bN4VIfH.1
Meeting ID: 960 9026 0834 Passcode: 374969