Abstract: Pretraining vision encoders with self-supervision (SSL) leads to stronger representations that excel across diverse downstream tasks. One of the key factors enabling self-supervision is extracting multiple views of the same scene to formulate either: 1) View-invariant pretraining (DINO, SimCLR, iBOT), where the objective is predicting the same representation for different views of the scene; or 2) Cross-view pretraining (cross-view Masked Autoencoders), where the objective is predicting missing parts of one view using other views. For extracting multiple views, view-invariant methods rely on a combination of handcrafted augmentations (random cropping, color jittering, gaussian blur, etc.) of the same image, whereas cross-view pretraining methods rely on image cropping or video frames. In this work, we present methods to effectively incorporate synthetic views from diffusion models into SSL training.
For view-invariant pretraining, we introduce Gen-SIS, a method that leverages the ability of diffusion models to generate interpolated images through interpolation in conditioning space. We introduce a disentanglement pretext task: disentangling two source images from an interpolated synthetic image. This disentanglement task, in addition to vanilla single-source generative augmentation for view extraction, improves visual pretraining of various view-invariant methods (DINO, SimCLR, iBOT).
For cross-view pretraining, we introduce CDG-MAE, a novel cross-view masked autoencoder (MAE) based method that uses diverse synthetic views generated from static images via an image-conditioned diffusion model to learn dense correspondences. We present a quantitative method to evaluate the local and global consistency of the generated views to choose the right diffusion model for cross-view pretraining. These generated views exhibit substantial changes in pose and perspective, providing a rich training signal that overcomes the limitations of video (expensive) and crop-based (less variation) methods. CDG-MAE substantially narrows the gap to video-based MAE methods on video label propagation tasks while maintaining the data advantages of image-only MAEs.

Speaker: Varun Belagali

Location: NCS 120
Zoom: https://stonybrook.zoom.us/j/93647452432?pwd=hZaX7LXCAD8KPHWYE1Afw2sDI3owpv.1
Abstract: DeepSeek-R1-Zero has shown that reinforcement learning (RL) at scale can directly enhance the reasoning capabilities of LLMs without supervised fine-tuning. In this work, we critically examine R1-Zero-like training by analyzing its two core components: base models and RL. We investigate a wide range of base models, including DeepSeek-V3-Base, to understand how pretraining characteristics influence RL performance. Our analysis reveals that DeepSeek-V3-Base already exhibit ''Aha moment'', while Qwen2.5 base models demonstrate strong reasoning capabilities even without prompt templates, suggesting potential pretraining biases. Additionally, we identify an optimization bias in Group Relative Policy Optimization (GRPO), which artificially increases response length (especially for incorrect outputs) during training. To address this, we introduce Dr. GRPO, an unbiased optimization method that improves token efficiency while maintaining reasoning performance. Leveraging these insights, we present a minimalist R1-Zero recipe that achieves 43.3% accuracy on AIME 2024 with a 7B base model, establishing a new state-of-the-art.

Speaker: Md. Saqib Hasan

Location: CS2311
The Art Department is hosting a guest artist exhibition, featuring the work of Young Maeng. The Opening Reception will be held on October 10th at 5 PM. Additionally, Young Maeng will be giving a talk on 'AI and Painting' on Oct 9 at 4:30 PM at the Future Histories Studio. Exhibition Location: Gallery Unbound, 3rd Floor, Staller Center, Stony Brook University
University Libraries Presents:
Join librarian Christine Fena for an interactive workshop that invites you to explore AI tools first hand, not just as users, but as critical investigators.
Through playful experimentation and collaborative discovery, you'll uncover inherent biases, probe algorithmic flaws, and gain a deeper understanding of AI's limitations and societal impacts.

RSVP on SBEngaged

Location: Melville Library, Central Reading Room, Lab B

Abstract: Large Language Models (LLMs) have revolutionized how people interact with knowledge, offering unprecedented opportunities to accelerate the pace of scientific discovery. In this talk, I will discuss my research on the synergy between LLMs and scientific knowledge--specifically how these models extract, induce, and verify knowledge to automate the research lifecycle. First, I will cover our work on improving knowledge extraction from vast scientific literature, focusing on enabling models to comprehend long documents in a cost-efficient and comprehensive manner. I will describe a novel paradigm for representing document-level structured information as question-answer pairs and how we address the challenges of long-context understanding by leveraging global context through retrieval-augmented modeling. Next, I present our pioneering work on using LLMs for new scientific hypothesis generation. We introduce a framework employing reinforcement learning with fine-grained reward modeling and adaptive controllers.
This approach balances novelty, feasibility, and effectiveness to generate inspiring and actionable research hypotheses. Finally, I will discuss work on the first LLM Scientist for machine learning research. I will demonstrate how LLMs can move beyond hypothesis generation to participate in the execution and validation of scientific hypotheses, ensuring that the discovered knowledge is not only innovative but also grounded and verified.

Bio: Xinya Du is a tenure-track assistant professor at UT Dallas Computer Science Department. He earned a Ph.D. degree from Cornell University and was a Postdoctoral Research Associate at the University of Illinois (UIUC). He has also worked at Microsoft Research, Google Research, and Allen Institute AI. His research is on large language models, deep learning, and their applications in science.His work has been published in leading NLP and ML conferences (ACL, ICLR, NeurIPS). His research has received multiple recognitions, including a Best Paper Award at AAAI AI for Research and a Best Poster Award at ICML AI for Science workshop. His work was included in the list of Most Influential ACL Papers and has been covered by major media like New Scientist. He was named a Spotlight Rising Star in Data Science by the University of Chicago and is the recipient of several prestigious awards, including the Amazon Research Award, Cisco Research Award, Open Philanthropy Award, and the NSF CAREER Award.

Location: NCS 120
Abstract:
Coarse grained (CG) models alleviate the drawbacks of all-atom simulations. The latter still pose challenges because they are computationally expensive and give access to limited spatiotemporal scales, despite the use of modern high-performance computing clusters. CG models ignore some of the atomistic degrees of freedom, leading to fewer interatomic interactions, hence less computing time. Introducing such models emphasizes the need to properly manage these multiple scales, by carefully deriving potentials and reconstructing conformations from their CG representations, usually with the help of Machine Learning. Following a bottom-up and force matching approach, we train a Physics-Informed Neural Network to extract the CG force field parameters from all-atom simulation data. We verify our approach by applying it to fibrin monomers to study multiple-fibrin polymerization in solution at the microsecond scale, after modifying the force field to incorporate further non-bonded interactions, not present in the training data. Access to these scales will allow us to study the effects of some of the molecules' components. Furthermore, we modify recent solutions in data-driven protein backmapping. Taking advantage of the developments in graph neural networks and variational inference, we introduce an intermediate step in the all-atom reconstruction of a molecule given its CG configuration, in an attempt to more accurately de-coarsen structures whose atom-to-CG-beads ratio is very high. The combined effect of our new forward and inverse coarse graining methodology will enable the in silico study of many phenomena that are highly dynamic and intrinsically multiscale.

Bio:
Georgios Kementzidis is a third year PhD student in the Department of Applied Mathematics and Statistics at Stony Brook University. His advisor is Dr. Yuefan Deng. His research interests lie at the intersection of Computational Science, molecular dynamics (MD) simulations, and Machine Learning (ML) applications to Computational Biophysics. He is particularly interested in coarse-graining and multi-scale simulations.

*Note: this seminar will be held in-person (food provided on a first-come, first serve basis) and online*

Join Zoom Meeting https://stonybrook.zoom.us/j/99510099036?pwd=EyowuLBGvUVLZDBlG6F6chkMICFOZ7.1
Meeting ID: 995 1009 9036
Passcode: 132419

Professor Petar M. Djuric, SUNY Distinguished Professor and Savitri Devi Bangaru Professor in Artificial Intelligence at Stony Brook University, has been selected as a plenary speaker at the upcoming 23rd IEEE Statistical Signal Processing Workshop (SSP 2025). The event will be held from June 8-11, 2025, in Edinburgh, Scotland, and is one of the premier international forums for the latest advances in statistical signal processing.

Professor Djuric's plenary talk, titled Quantifying causal relationships: Dynamic strengths, attributions, and confounders, will take place on June 10 from 9:00 AM to 10:00 AM EST. His presentation addresses foundational challenges in data-driven causality, proposing novel methodologies for quantifying causal strength in both static and dynamic systems, with special attention to latent confounders and attribution analysis.

This work has broad implications across disciplines including healthcare, economics, and climate science--areas where causal understanding drives critical decisions and innovations.

Professor Djuric has been a long-standing leader in the fields of machine learning and signal and information processing. After receiving his Ph.D. from the University of Rhode Island, he joined the faculty at Stony Brook University, where he served as Chair of the Department of Electrical and Computer Engineering from 2016 to 2023. He is also the founding Editor-in-Chief of the IEEE Transactions on Signal and Information Processing Over Networks and a Fellow of IEEE, EURASIP, AAIA, and AIIA.

Early bird registration for the workshop is open until April 30, 2025. For more information, visit the official SSP 2025 website.

AI is everywhere and so are the privacy concerns that come with it. At its core, the most common forms of AI we use today are online digital services and thus inherit the usual privacy risks. We'll take a look at indirect prompt injection- a technique that can trick AI tools into revealing or extracting private information as well as techniques being used in academic contexts to manipulate systems and even mislead researchers.

Register here for the online session.