Digital Commons Network^™

Nexus: Network Exploration For Exploiting Unsafe Sequences In Multi-Turn Llm Jailbreaks, Javad Rafiei Asl, Sidhant Narula, Mohammad Ghasemigol, Eduardo Blanco, Daniel Takabi

Computer Science Faculty Publications

Large Language Models (LLMs) have revolutionized natural language processing, yet remain vulnerable to jailbreak attacks—particularly multi-turn jailbreaks that distribute malicious intent across benign exchanges, thereby bypassing alignment mechanisms. Existing approaches often suffer from limited exploration of the adversarial space, rely on hand-crafted heuristics, or lack systematic query refinement. We propose NEXUS (Network Exploration for eXploiting Unsafe Sequences), a modular framework for constructing, refining, and executing optimized multi-turn attacks. NEXUS comprises: (1) ThoughtNet, which hierarchically expands a harmful intent into a structured semantic network of topics, entities, and query chains; (2) a feedback-driven Simulator that iteratively refines and prunes these chains …

A Proposed Ehrenfeucht-Fraïssé Game Model For Natural Language Processing Generative Adversarial Networks, Don Li

Anthós

Large Language Models (LLM’s) (e.g., ChatGPT) constitute both a significant research area and commercial application of AI. Current major LLM’s are built on Generative Pre-Trained Transformer (GPT) neural network architecture to perform natural language processing (NLP) tasks. Generative Adversarial Network (GAN) is another popular neural network architecture, which leverages a zero-sum game between constituent neural networks within the architecture to train the GAN, and is widely used for visual data applications. This article proposes a new GAN architecture for NLP: an EF-GAN whose underlying algorithm uses Ehrenfeucht–Fraïssé (EF) games, a game-theoretic approach from model theory to determine elementary equivalence of …

Editorial Insights: Reflections On The Volume And Charge To The Field, Heather Howell, Meghan Shaughnessy, Barbara Stengal, Carrie Wilkerson Lee, Liza Bondurant, Bima Sapkota, Gregory Benoit, Yvonne Lai

School of Mathematical & Statistical Sciences Faculty Publications

In this, the final chapter of Section 3, the authors document an interview that took place in June of 2024, as the editorial team wrapped up reviews of the chapters submitted to this volume. The interview was led by two of the advisory board members, Meghan Shaughnessy and Barbara Stengal, and included the editorial group, Heather Howell, Carrie Wilkerson Lee, Liza Bondurant, and Bima Sapkota, and other members of the advisory board, Greg Benoit and Yvonne Lai. The motivation to create this interview-based chapter was to capture some of the authors' own learning and reflection as they brought the volume …

Opinion Mining On Offshore Wind Energy For Environmental Engineering, Isabele Bittencourt, Aparna S. Varde, Pankaj Lal

School of Computing Faculty Scholarship and Creative Works

Renewable energy sources are vital to help mitigate the effects of climate change, and reducing the carbon dioxide emissions of fossil fuels, e.g. the state of New Jersey has a goal of producing 100% clean energy by 2050. However, the plans for offshore wind energy by the shore of the state still brings much controversy between residents due to the wind farms’ impact on wildlife, coastline, and the people’s view from the beaches. In this context, we perform sentiment analysis on social media data to investigate people’s opinions and concerns regarding offshore wind energy. We adapt 3 machine learning models, …

Leveraging Large Language Models To Create Learner Personas For Training Design, N. Lovett, M. Yang, K. Herman, B. Li, M. Sosonkina, W. Purwanto, P. Jiang, M. H. Wu

Electrical & Computer Engineering Faculty Publications

This case examines the innovative use of Large Language Models (LLMs) to generate learner personas for developing learner-centered cybersecurity training materials when direct access to initial learner data is not available. The team developed a nine-stage iterative process for creating and refining AI-generated personas to address this constraint, integrating ethical review, stakeholder feedback, and action research principles. The process expanded upon Kouprie and Visser’s (2009) empathic design framework to ensure cultural responsiveness and mitigate potential biases in LLM outputs. Through multiple refinement cycles, initial generic personas evolved into detailed, context-rich archetypes which informed the development of effective and context-responsive training …

Symp25s: Can Llm Detect Dementia?, Rishank Singh, Youxiang Zhu, Xiaohui Liang, John A. Batsis, Caroline Summerour

Paul English Applied Artificial Intelligence (AI) Institute Publications

High Cost of Traditional Screening: Formal cognitive assessments for dementia are resource-intensive and not easily accessible for large-scale screening. Speech-Based Alternatives: Existing speech-based methods (e.g., picture description, telephone interviews) aim to address this but have limitations. Lack of Natural Dialogue: These conventional approaches often use rigid, repetitive prompts and do not simulate real conversations. Engagement Issues: Repetition and lack of conversational depth can reduce engagement and affect the accuracy of responses over time. Untapped Potential of LLMs: Large language models (LLMs) are capable of generating natural, coherent, and adaptive dialogue. Research Gap: The application of LLMs for dementia detection through …

Optimizing Ai Language Models: A Study Of Chatgpt-4 Vs. Chatgpt-4o, Md Nurul Absar Siddiky, Muhammad Enayetur Rahman, Md Fayaz Bin Hossen, Muhammad Rezaur Rahman, Md. Shahadat Jaman

Electrical & Computer Engineering Faculty Publications

This paper presents a comparative analysis of OpenAI's GPT-4 and its optimized variant, GPT-4o, focusing on their architectural differences, performance, and real-world applications. GPT-4, built upon the Transformer architecture, has set new standards in natural language processing (NLP) with its capacity to generate coherent and contextually relevant text across a wide range of tasks. However, its computational demands, requiring substantial hardware resources, make it less accessible for smaller organizations and real-time applications. In contrast, GPT-4o addresses these challenges by incorporating optimizations such as model compression, parameter pruning, and memory-efficient computation, allowing it to deliver similar performance with significantly lower computational …

Performance Of Standard Medical Mllms On Ecg Image Data, Prisha Anil

Masters Theses

This work presents a structured benchmarking study of multimodal large language models (MLLMs) applied to electrocardiogram (ECG) interpretation tasks. We evaluate three representative architectures: MedGemma, HuatuoGPT-Vision, and LLaVA-Med, across progressive experimental stages involving text-only structured prompt normalization, text–image fusion with ECG plots, and full multimodal fusion incorporating time-series signals. A standardized five-section cardiology prompt was designed to enforce consistent output structure and SCP-code alignment, enabling reproducible metric computation across models. Quantitative evaluation using BERTScore, token-level F1, and diagnostic accuracy demonstrates that HuatuoGPT-Vision achieves the highest semantic and diagnostic alignment, while MedGemma exhibits superior formatting stability and reproducibility. In contrast, LLaVA-Med …

Relationship Extraction Using Retrieval Augmented Generation For Biomedical Dataset, Jannat Jannat

Theses and Dissertations

With the increasing number of structured and unstructured data, obtaining reliable information effectively has become crucial. In the biomedical domain, extracting information from the scientific papers is crucial in order to stay up-to-date with accurate information, given the increased pace by which new research studies are published. This work focuses on identifying relationships between entities that are extracted from the abstracts and titles of biomedical research papers. In this work, we developed a Retrieval Augmented Generation (RAG) based system to automatically identify relations between biomedical entities. We evaluate multiple open source Large Language Models (LLMs) and the number of examples …

“Lost-In-The-Later”: Framework For Quantifying Contextual Grounding In Large Language Models, Yufei Tao, Adam Hiatt, Rahul Seetharaman, Ameeta Agrawal

Computer Science Faculty Publications and Presentations

Large language models (LLMs) are capable of leveraging both contextual and parametric knowledge but how they prioritize and integrate these sources remains underexplored. We introduce CoPE, a novel framework for systematically quantifying contextual grounding in LLMs. CoPE distinguishes between contextual knowledge (CK) and parametric knowledge (PK), enabling fine-grained attribution across languages and tasks. Using our newly created MultiWikiAtomic dataset in English, Spanish, and Danish, we analyze how LLMs integrate context, prioritize information, and incorporate PK in open-ended question answering. We find that across models and languages, only around 50 to 76 percent of outputs are grounded in the given context, …

Scitopic: Enhancing Topic Discovery In Scientific Literature Through Advanced Llm, Pengjiang Li, Zaitian Wang, Xinhao Zhang, Ran Zhang, Lu Jiang, Pengfei Wang, Yuanchun Zhou

Computer Science Faculty Publications and Presentations

Topic discovery in scientific literature provides valuable insights for researchers to identify emerging trends and explore new avenues for investigation, facilitating easier scientific information retrieval. Many machine learning methods, particularly deep embedding techniques, have been applied to discover research topics. However, most existing topic discovery methods rely on word embedding to capture the semantics and lack a comprehensive understanding of scientific publications, struggling with complex, high-dimensional text relationships. Inspired by the exceptional comprehension of textual information by large language models (LLMs), we propose an advanced topic discovery method enhanced by LLMs to improve scientific topic identification, namely SciTopic. Specifically, we …

A Systematic Review On Mental Health Chatbots: Trends, Design Principles, Evaluation Methods, And Future Research Agenda, Mohammad Amin Kuhail, Ons Al-Shamaileh, Shahbano Farooq, Hana Shahin, Fatema Abdelzaher, Justin Thomas

All Works

Recent decades have witnessed a rise in the prevalence of common mental health problems coupled with an increasing demand for talk‐based psychological therapies. This has coincided with the rise of mental health chatbots (MHCs), a corollary of which is the use of MHCs, AI‐driven, conversational agents, to provide interactive support, guidance, and therapeutic engagement for persons experiencing mental health challenges. There has been a growing research interest in MHCs, and this study provides a much‐needed systematic review that examines this expanding research literature, identifying themes, trends, and areas worthy of further exploration. This review identified and systematically explored 97 published …

Ai Characterisations And Their Legal Implications, Jerrold Tsin Howe Soh

Research Collection Yong Pung How School Of Law

This chapter examines the difficult legal characterisation problems that artificially intelligent systems raise and explores how different characterisations of artificial intelligence (AI) shape practical legal outcomes. Three reasons are offered for the legal difficulty with characterising AI. First, answers to characterisation problems are inherently subjective and perspective-driven, particularly when the subject is an intangible technological system. Second, AI technology is especially difficult to define since the field typically proceeds on inexact anthropomorphic metaphors. Third, AI characterisation problems raise difficult sub-problems, particularly in determining how autonomous an AI system is. The chapter thus argues that a range of plausible AI characterisations …

Towards Human Explainable Digital Forensics: Generating Human Interpretable Evidence For Semantic Understanding In Manipulated Images And Text, Yuwei Chen

Electronic Theses & Dissertations (2024 - present)

Detecting and characterizing manipulations in digital media continues to pose a significant challenge within the field of digital forensics. Despite notable advancements, the discipline often remains in a reactive stance against emerging threats. Current state-of-the-art methods, typically evaluated within academic settings, fails to mirror the complexities of real-world disinformation scenarios. These methods generally prioritize high performance based on quantitative metrics, yet they demonstrate a considerable dependency on training data and lack adaptability to new novel attack signatures. With the rapid evolution of attack methodologies, the dependency on highly accurate models that do not generalize or adapt well to unseen threats …

Pushing The Boundaries Of Large Language Models: Innovations And Limitations In Nlp, Finance, And Mathematics, A M Muntasir Rahman

Dissertations

Large Language Models (LLMs) have emerged as transformative tools across a spectrum of domains, yet their practical deployment reveals a blend of remarkable potential and notable limitations. This research explores innovative methodologies to extend the capabilities of LLMs while addressing critical challenges in their evaluation and application. By leveraging rule-based approaches, the in-context learning capabilities of LLMs, and human-in-the-loop validation across three focused studies, this research introduces robust strategies for dataset synthesis, model enhancement, and model assessment in three distinct domains: natural language processing, financial sentiment analysis, and mathematical reasoning

The first study proposes an efficient data augmentation framework, EASE, …

Composite Volume Learning Media Based On Augmented Reality To Enhance Students’ Visual-Spatial Intelligence, Ria Kurniawati, Susiswo Susiswo, Syamsul Hadi

Jurnal Pendidikan Sains

Augmented Reality (AR) technology, a three-dimensional visual tool that integrates virtual objects or environments into real-world displays in real time (Azuma et al., 2001), is particularly useful for grasping complex concepts such as the composite volume of solids, which often confuses students when they need to decompose solids into their individual components. While previous research has predominantly focused on the introduction of solids, spatial nets, and the volume of singular shapes, this study aims to develop composite volume learning media based on AR to enhance students’ visual-spatial intelligence in a valid, practical, and effective manner, aligning with their concrete operational …

A Novel Format For Math Circles: Circles With No Center, Frederick A. Peck, Ian P. Renga, Nick Grener, David R. Erickson, Matt B. Roscoe, Ke Wu

Journal of Math Circles

This paper shares a novel format for Math Circles, which we call “circles with no center.” Circles with no center do not have a traditional facilitator. The problem or activity is novel to everyone in the room. All members engage in mathematical inquiry, and everyone takes responsibility for running the gathering. The “no center” format has benefits for participants and organizers. For participants, the format offers more freedom and more responsibility. For organizers, the format is easier to organize, more fun, and involves less pressure. Our assessment of a 3-day retreat for teachers suggests participants may prefer a circle with …

Narrative Engineering: Tools, Computational Structure, And Impact Of Stories, Michael A. Debuse

Theses and Dissertations

Computational Linguistics has a long history of applying mathematics to the grammatical and syntactic structure of language; however, applying math to the more complex aspects of language, such as narrative, plot, scenes, character relations, causation, etc. remains a difficult topic. The goal of my research is to bridge the narrative humanities with mathematics, to computationally grasp at these difficult topic, and help develop the field of Narrative Engineering. I view narrative and story with the same mathematical scrutiny as other engineering fields, to take the creativity and fluidity of story and encode it in mathematical representations that have meaning beyond …

Measuring And Improving Api Usability And Quality: A Comprehensive Framework And Empirical Study, Sultan Alanazy

Computer Science and Engineering Theses and Dissertations

Cloud computing provides on-demand access to flexible computing resources, enabling rapid application deployment without substantial infrastructure investment. Application Programming Interfaces (APIs) play an important role in ensuring the success of cloud applications. The primary users of APIs are the extensive community of application programmers who search, read, and understand APIs before integrating them into their applications or systems. In addition, developers often turn to online API support when seeking help. Problems in such support can result in incorrect API usage and integration problems. There is an urgent need to measure API usability and support issues to identify, characterize, and assess …

Enhancing Graph Neural Networks By Editing Graphs, Jiayu Li

Dissertations - ALL

Graphs are pervasive in both the natural world and various domains of science and engineering. Numerous advanced classifiers, such as Graph Neural Networks (GNNs), have been developed to perform node classification on these graphs. However, as graphs become denser with an increasing number of edges, GNNs often suffer from suboptimal generalization performance due to the presence of task-irrelevant connections. These redundant connections can introduce noise, consume excessive computational resources, and degrade performance. Identifying and preserving critical connections in large-scale graphs, while pruning unnecessary ones, is crucial for enhancing the efficiency and accuracy of GNNs in node classification, particularly for GCNs. …

Privacy Protection In Machine Learning Via Exploitation Of Constrained Adversarial Evasion, Brian Testa

Dissertations - ALL

Deep neural networks are extensively applied to real-world tasks. In many cases, the data feeding these tasks are human generated content, which emphasizes the criticality of privacy and data protection. The diverse modalities of this user data provide fertile ground for 3rd parties to monetize a user’s data. This work considers two such modalities: user speech when interacting with smart speaker voice assistants (VAs) and images shared with online service providers. In these cases, the user would like to support some form of machine learning (ML) inference without allowing others. A user interacting with a smart speaker would like the …

Overcoming Obstacles To Interdisciplinary Research: Empirical Insights And Strategies [Peer-Reviewed Journal], Cristo Leon Ph.D., James Lipuma Ph.D.

STEM for Success Resources

Explore how to overcome obstacles to interdisciplinary research with strategies for transdisciplinary communication, breaking silos, and fostering collaboration. Read insights from Cristo Leon and James Lipuma in this impactful study.

Navigating Transdisciplinary Communication: A Graduate Student's Perspective, Sirimuvva Pathikonda, James Lipuma, Cristo Leon

STEM for Success Resources

Discover "Navigating Transdisciplinary Communication: A Graduate Student’s Perspective," a comprehensive exploration of how diverse disciplines converge to tackle complex challenges. This study, presented at WMSCI 2024, employs a Constructivist lens to highlight the co-construction of knowledge, innovative teamwork tools, and practical strategies for research in STEM and beyond. Co-authored by Sirimuvva Pathikonda, MEM, and James Lipuma, Ph.D., the paper underscores the value of Transdisciplinary Communication (TDC) in fostering collaboration, innovation, and holistic problem-solving across fields. Read the full article here: https://doi.org/10.54808/JSCI.22.06.151.

Improving Clinical Information Extraction From Electronic Health Records: Leveraging Large Language Models And Evaluating Their Outputs, Kriti Bhattarai

McKelvey School of Engineering Graduate Student Theses & Dissertations

Accurate extraction of clinical entities and phenotypes from unstructured electronic health record (EHR) text is crucial for various clinical research tasks, including cohort identification, tracking temporal patterns in disease progression and deciding treatment course. However, this task remains challenging due to the complexity and ambiguity of medical language. This dissertation explores the application of advanced generative pre-trained transformer (GPT) models, such as GPT-4, GPT-3.5-turbo, Llama-3.1, Llama-3 and Flan-T5, for clinical entity and phenotype extraction from EHRs. Building upon these findings, this dissertation also investigates a hybrid approach where integration of external knowledge sources, such as Unified Medical Language System (UMLS) …

Implication Of Generative Ai On Education And Research, Riddhi Gupta

The Journal of Purdue Undergraduate Research

No abstract provided.

A Meteorological Investigation Of A Tornado- Producing, “Hybrid” Supercell-Multicell Storm Near Delphi, Indiana, On May 8, 2023, Evelyn Rose K. Girardi

The Journal of Purdue Undergraduate Research

No abstract provided.

Embodied Cognition And Teaching Stem: Tenets To Explain And Enhance A Middle School Science Project, Jonathan E. Roberts, Joshua L. Williams, Robert A. Hodgdon, Caroline Payne, Gabriela Ruiz Emmanuelli

Current Issues in Middle Level Education

STEM pedagogy is a popular area for research and discussion. The empirical literature on this topic supports a variety of approaches (e.g., inquiry-based) to help instructors spark student interest create an effective learning environment, and foster long-term retention and transfer of information related to STEM topics. One potential approach is to use tenets of embodied cognition, in which behavior and cognition emerge out of real-time sensorimotor behavior of the individual situated within a particular context, to implement strategies to enhance STEM pedagogy. We applied the six lessons of embodied cognition to understand the beneficial aspects of a middle-school project on …

Utilizing Large Language Models To Synthesize Product Desirability Datasets, John D. Hastings, Sherri Weitl-Harms, Joseph Doty, Zachary L. Myers, Warren Thompson

Research & Publications

This research explores the application of large language models (LLMs) to generate synthetic datasets for Product Desirability Toolkit (PDT) testing, a key component in evaluating user sentiment and product experience. Utilizing gpt-4o-mini, a cost-effective alternative to larger commercial LLMs, three methods, Word+Review, Review+Word, and Supply-Word, were each used to synthesize 1000 product reviews. The generated datasets were assessed for sentiment alignment, textual diversity, and data generation cost. Results demonstrated high sentiment alignment across all methods, with Pearson correlations ranging from 0.93 to 0.97. Supply-Word exhibited the highest diversity and coverage of PDT terms, although with increased generation costs. Despite minor …

Advanced Models For Linking Process In Data Washing Machine, Bushra Sajid

Theses and Dissertations

Entity Resolution (ER) is a critical process in data integration and quality improvement that identifies and links multiple records referring to the same real-world entity. As data volumes and heterogeneity increase, traditional ER methods face new challenges, prompting research into more advanced techniques. The Proof-of-Concept Data Washing Machine (DWM), developed under the NSF DART Data Life Cycle and Curation research theme, aims to automatically detect and correct data quality errors through unsupervised entity resolution. Recent research focuses on enhancing DWM's effectiveness by replacing rule-based methods with machine learning and deep learning approaches, particularly in the linking process. Deep learning models, …

La Creatividad En Peligro: Como La Inteligencia Artificial Es Un Reto Para Los Artistas., Nathaly Cisneros

Capstones

Los artistas digitales han creado obras maestras que nos han dejado sin aliento con sus pinceles digitales, lápices y pinturas. Desde retratos que parecen saltar de la pantalla hasta paisajes que nos transportan a mundos desconocidos, su arte ha sido una fuente constante de inspiración.

Pero en los últimos años, una nueva fuerza ha comenzado a cambiar el juego. La inteligencia artificial ha estado avanzando a pasos agigantados y ahora se perfila como una amenaza para el futuro de los artistas digitales. ¿Qué significa esto para el arte y la creatividad?

Link: https://docs.google.com/document/d/1xe8UxDMekX_SwiIppyt_JppK8M-lB-YWNWGyeyShlJM/edit?usp=sharing