Title:Deep Contextual Modeling for Natural Language Understanding, Generation, and Grounding Zoom instructions: Join Zoom Meeting https://stonybrook.zoom.us/j/645050299?pwd=TVJVRkc3dlhxdDF5d00xWGlDQkovZz09 Meeting ID: 645 050 299 Password: 810247 One tap mobile +16468769923,,645050299#,,#,810247# US (New York) +13126266799,,645050299#,,#,810247# US (Chicago) Dial by your location +1 646 876 9923 US (New York) +1 312 626 6799 US (Chicago) +1 301 715 8592 US +1 346 248 7799 US (Houston) +1 408 638 0968 US (San Jose) +1 669 900 6833 US (San Jose) +1 253 215 8782 US Meeting ID: 645 050 299 Password: 810247 Find your local number: https://stonybrook.zoom.us/u/aemTiJMXu6 Abstract: Natural language is a fundamental form of information and communication. In both human-human and human-computer communication, people reason about the context of text and world state to understand language and produce language response. In this talk, I present several deep neural network based systems that first understand the meaning of language grounded in various contexts where the language is used, and then generate effective language responses in different forms for information access and human-computer communication. First, I will introduce Speaker Interaction RNNs for addressee and response selection in multi-party conversations based on explicit representations for different discourse participants. Then, I will present a text summarization approach for generating email subject lines by optimizing quality scores in a reinforcement learning framework. Finally, I will show an editing-based multi-turn SQL query generation system towards intelligent natural language interfaces to databases. Bio:Rui Zhang is a final year Ph.D. student at Yale University advised by Professor Dragomir Radev. His research interest lies in various natural language processing problems in understanding, generation, and grounding. He has been working on (1) End-to-End Neural Modeling for Entities, Sentences, Documents, and Multi-party Multi-turn Dialogues, (2) Text Summarization for Emails, News, and Scientific Articles, (3) Cross-lingual Information Retrieval for Low-Resource Languages, (4) Context-Dependent Text-to-SQL Semantic Parsing in Human-Computer Interaction. Rui Zhang has published papers and served as Program Committee members at top-tier NLP and AI conferences including ACL, NAACL, EMNLP, AAAI, CoNLL. During his Ph.D., He has done research internships at IBM Thomas J. Watson Research Center, Grammarly Research, and Google AI. He was a graduate student at the University of Michigan and got his bachelor's degrees at both the University of Michigan and Shanghai Jiao Tong University from the UM-SJTU Joint Institute.
Synthetic Dreams and Ghost Machines

Dr. Steven Skiena will join the Cinema for a presentation exploring the rapidly evolving worlds of artificial intelligence and robotics.

Dr. Skiena is Distinguished Teaching Professor of Computer Science and Associate Director of the AI Innovation Institute at Stony Brook University. A leading researcher in data science and algorithms, he is the author of several influential books on AI and computation, including The Algorithm Design Manual and The Data Science Design Manual.

The presentation will be followed by a screening of Ex Machina, Alex Garland's provocative science-fiction thriller exploring the seductive and unsettling boundaries between human consciousness and artificial intelligence.

Looking to learn about a new topic or skill? Look no further! Gemini's Guided Learning feature acts as your own personal tutor, teaching you about a particular subject through an engaging back and forth conversation. This AI tool helps users develop their knowledge and skills on a wide variety of topics, acting as a patient mentor, breaking down complex topics step-by-step. This session will take place on 2/24 at 11 AM. Please register using the link below!
https://stonybrookuniversity.co1.qualtrics.com/jfe/form/SV_a9PVlBw0E1Bal1A?
The GE Vernova Advanced Research Center invites you to the 2026 AI EDGE Symposium.

This collaborative and hands-on learning experience connects 300+ industry thought leaders in the fields of AI, Edge, Robotics, Cybersecurity, and Controls & Optimization.


Attendees will learn about the latest challenges and innovations with luminaries from key government agencies, industry customers, and technology partners as they discuss the latest advancements and trends in these fields.

Register here.

Submit an abstract celebrating research, new discoveries and achievements in medicine and science!

We encourage faculty, nurse practitioners, post-doctoral fellows, fellows, residents, medical students, graduate students and undergraduate students to submit an abstract. Original research, case reports and case series are welcome.

Abstract submission deadline: FEBRUARY 7, 2025

For more details, visit here.

Abstract: Olfaction plays an essential role in everyday life. Anosmia, the loss of the sense of smell, affects approximately 20% of adults globally and its prevalence rises with age. This can lead to malnutrition, reduced ability to detect environmental hazards, and high rates of depression. The olfactory epithelium (OE) is a specialized tissue deep within the nasal cavity that primarily contains olfactory sensory neurons (OSNs), which are crucial cells that detect odorants and relay information to the olfactory neural pathway. Current clinical assessments for anosmia cannot discern whether the cause lies within the OSNs (sensorineural) or further down the pathway (central). The ability to image the OE would help assess the health of OSNs, distinguish between sensorineural and central causes of anosmia, and inform treatments. However, the OE is a small region of tissue that is non-uniform and difficult to distinguish from its surrounding respiratory epithelia (RE). This work explores the use of an FDA-approved dye, Indocyanine Green (ICG), to highlight tissue structure differences and help distinguish between OE and RE. We found that machine learning models using texture analysis information were able to classify images of the two tissue types correctly with ~80% accuracy, while a convolutional neural network was able to classify the images directly with ~90% accuracy. This advancement will enable accurate identification and mapping of the spatial distribution of OSNs within the nasal cavity, enhancing studies on olfactory degeneration and regeneration.


Bio: Skylar Suarez has a Bachelor of Science degree in Computer Science and Engineering from the University of South Florida. She spent nearly a decade as a software and hardware engineer, then decided to follow my dream of being involved in medicine and becoming a professor. She is a PhD Candidate at the Department of Biomedical Engineering at the University of Colorado Anschutz Medical Campus, where she also teaches courses in biological data analysis and machine learning. She specializes in neural engineering, particularly in the development and design of neurosurgical tools for assessing neurodegenerative disorders.


Location: New Computer Science, Room 220 OR Join by Zoom
Educational objectives:
  1. Techniques involving data driven inferences from multi-omics data
  2. Interdisciplinary approaches to uncover aspects of tumor biology
  3. Computational genomics method development, benchmarking, and validation


Speaker: Dr. Subhajyoti De, PhDProfessor, Cancer SystemsBiology and Pathology,Rutgers Cancer Institute,The State University of New Jersey

Location: MART Building, Room 7M-0602 (7th Floor)

Remote Access: https://stonybrook.zoom.us/j/95617197636?pwd=KytzZ2pVRG9SZGpKZUtpNXJISjNjZz09
Meeting ID: 95617197636 | Passcode: 924293
The Natural Language Processing Reading Group at Stony Brook University meets weekly to discuss recent research papers in NLP and related fields.
Join the Google Group here.


Why LLM Safety Guardrails Collapse After Fine-tuning: A Similarity Analysis Between Alignment and Fine-tuning Datasets

Abstract: Recent advancements in large language models (LLMs) have underscored their vulnerability to safety alignment jailbreaks, particularly when subjected to downstream fine-tuning. However, existing mitigation strategies primarily focus on reactively addressing jailbreak incidents after safety guardrails have been compromised, removing harmful gradients during fine-tuning, or continuously reinforcing safety alignment throughout fine-tuning. As such, they tend to overlook a critical upstream factor: the role of the original safety-alignment data. This paper therefore investigates the degradation of safety guardrails through the lens of representation similarity between upstream alignment datasets and downstream fine-tuning tasks. Our experiments demonstrate that high similarity between these datasets significantly weakens safety guardrails, making models more susceptible to jailbreaks. Conversely, low similarity between these two types of datasets yields substantially more robust models and thus reduces harmfulness score by up to 10.33%. By highlighting the importance of upstream dataset design in the building of durable safety guardrails and reducing real-world vulnerability to jailbreak attacks, these findings offer actionable insights for fine-tuning service providers to prioritize upstream models with low jailbreak risk.

Location: NCS 220

Zoom Link: https://stonybrook.zoom.us/j/92942833294?pwd=927683