First, the dissertation focuses on uncovering cognitive styles or thinking patterns manifesting in language. We demonstrate that modeling language at deeper semantic levels such as discourse relations, can unveil latent psychological states and traits, including cognitive styles that influence both mental health and behavior. Introducing a novel blend of transfer and active learning, we efficiently curated a new set of linguistic data on cognitive styles like dissonance. This approach allows for more precise measurement when dealing with rare-classes and low-resource tasks. As a second contribution, effective validation methods are introduced to language-based assessments of the underlying cognitive styles. Controlled behavioral experiments and online studies show that cognitive styles detected through linguistic signals reliably predict real-world behaviors such as decision-making and engagement with extremist communities, both at the individual and community levels, sometimes months in advance
The research further moves beyond traditional measurement tools like questionnaires and expert judgments, which rely on Classical Test Theory, by establishing that language-based assessments more closely approximate true psychological states. The mechanisms by which these assessments outperform standard tools are explained, highlighting their predictive power for behaviors linked to underlying traits. Finally, a more sophisticated approach is explored by modeling psychological outcomes with Item Response Theory (IRT), an improvement over Classical Test Theory. Adaptive language-based assessments are introduced, showing that targeted, adaptive testing based on latent IRT scores can efficiently and accurately capture multiple psychological dimensions.
Taken together, these contributions argue for a shift towards language-based psychological assessments. By integrating deeper discourse-level semantics with measurement theory, this dissertation charts a path towards truer scores of mental states: ones that are more precise, and reflective of the complexity of human cognition and emotions.
Speaker: Vasudha Varadarajan
https://stonybrook.zoom.us/j/
The next AI Institute seminar speaker will be Chao Chen of Biomedical Informatics, on Monday November 29 at noon via zoom:
https://stonybrook.zoom.us/j/
He will be talking on the Detection of Trojan Attacks to Deep Neural Networks - A Topological Perspective, with his abstract and bio below.
Abstract: Deep neural networks are known to have security issues. One particular threat is the Trojan attack. It occurs when the attackers stealthily manipulate the model's behavior through Trojaned training samples, i.e., samples with special trigger injected and labels altered. To identify a Trojaned model at deployment is challenging, due to limited access to the training data. We propose to identify Trojaned neural networks using methods from topological data analysis. In particular, we propose to (1) inspect high-order topological features of the neuron interactions and (2) reverse engineer the injected triggers using a topological loss. These approaches take different angles and reveal insights into the behavior of neural networks when their strong memorialization power is exploited maliciously. The work has been accepted to NeurIPS'21. I will also briefly mention other research directions from my group, including incorporating topological information into deep image analysis, topology-inspired graph neural networks, and robust training of neural networks with label noise. These works have been published in ICLR, ICML, NeurIPS, ECCV, ICCV and AAAI in recent years.
Bio: Dr. Chao Chen is an assistant professor of Biomedical Informatics at Stony Brook University. His research interests span topological data analysis (TDA), machine learning and biomedical image analysis. He develops principled learning methods inspired by the theory from TDA, such as persistent homology and discrete Morse theory. These methods address problems in biomedical image analysis, robust machine learning, and graph neural networks from a unique topological view. His research results have been published in major machine learning, computer vision, and medical image analysis conferences. He is serving as an area chair for MICCAI, AAAI, CVPR and NeurIPS.
Learn about features in the + icon in Gemini.
The plus icon in the Gemini prompt gives you access to different capabilities. We'll demo a few of these features to show how they can help you in your daily work. We'll look at how guided learning, writing, visual creation, and research can support your work.
The Vedanta Forum is devoted to one of humanity's oldest and most profound pursuits -- thinking. Thinking about who we truly are: the one that remains constant through childhood and old age, through waking, dream, and deep sleep. Thinking about the source and cause of creation, and its relationship to what inheres in us.
Across history, such thinking, both meditative and scientific, has been aimed at these questions. The ancient Upanishads proclaimed, Tat Tvam Asi -- Thou Art That -- revealing the non-dual identity of the individual and the ultimate reality. Centuries later, modern scientists such as Schrödinger and Bohr echoed similar intuitions about the unity of existence.
Over time, many philosophical approaches, traditions, and interpretive schools have arisen from such inquiry, each offering unique perspectives. The Forum will:
Focus on universal approaches and traditions and examine their teachings,
Foster comparative studies, and
Explore the practical benefits to society from such thinking,
through scholarly studies, dialogue, and debate also promoting accessibility to all qualified seekers. Additionally, the Forum will explore how these reflections can enrich life, education, and even technology.
Location: NCS 120 (New Computer Science), Engineering Dr, Stony Brook, NY 11794.
The program is available at: https://www.vedantaforum.org/
I will show how neural networks, trained to predict the eddy diffusivity profile from high-fidelity yet computationally expensive turbulence schemes, enhance the vertical mixing scheme in the climate model. These networks replace ad hoc components while maintaining the conservation principles of the standard ocean model equations. The enhanced scheme outperforms its predecessor by reducing biases in the mixed-layer depth and modestly improving tropical upper-ocean stratification in ocean-only global simulations. Furthermore, simplified equations that can replace the neural networks show similar improvements but with lower computational cost and better interpretability. They point to structural deficiencies in the baseline parameterization. This work is one of the first successful applications of machine learning to improve a sub-grid parameterization of turbulent mixing in ocean climate models.
IACS Seminar Speaker: Aakash Sane, Princeton University
Location: IACS Seminar Room or Zoom
Join Zoom Meeting: https://stonybrook.zoom.us/j/97764942108?pwd=MzCWupCe3L9mKdrgfO2bJg3GBbvXuf.1
Meeting ID: 977 6494 2108
Passcode: 519324
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
Zoom link to follow.
Abstract: Document classifiers have been widely applied in solving health-related issues, such as suicide prevention, flu vaccination surveillance and disease diagnosis. However, document metadata including time, gender, age and location has an enormous impact on robustness of
document classifiers. Language varies across the metadata bringing both challenges and opportunities to build reliable document classifiers. For example, online written language changes over time, and males and females express opinions differently. This talk describes how to use domain adaptation to integrate temporal and user demographic factors into document classifiers. By adapting knowledge of how language varies across the metadata, models can learn generalized representations of language through the metadata-invariant embeddings.
This approach will lead to metadata-adapted document classifiers and can also extend to personalize classification models by user embedding.
Bio: Xiaolei Huang is a 4th-year PhD candidate in Information Science at the University of Colorado, Boulder. He is currently a visiting scholar at the Johns Hopkins University. His research interests are in Natural Language Processing, Machine Learning and Public Health. Particularly, he focuses on domain adaptation, cross-lingual transfer learning, user modeling and fairness.
Location: SAC Room 304