Digital Commons Network^™

De-Identification Is Not Enough: A Comparison Between De-Identified And Synthetic Clinical Notes, Atiquer Rahman Sarkar, Yao-Shun Chuang, Noman Mohammed, Xiaoqian Jiang

Faculty, Staff and Student Publications

For sharing privacy-sensitive data, de-identification is commonly regarded as adequate for safeguarding privacy. Synthetic data is also being considered as a privacy-preserving alternative. Recent successes with numerical and tabular data generative models and the breakthroughs in large generative language models raise the question of whether synthetically generated clinical notes could be a viable alternative to real notes for research purposes. In this work, we demonstrated that (i) de-identification of real clinical notes does not protect records against a membership inference attack, (ii) proposed a novel approach to generate synthetic clinical notes using the current state-of-the-art large language models, (iii) evaluated …

Beyond Climate In Placencia, Belize: Navigating Environmental Change In The Anthropocene, Rory E. Mckenna

USF Tampa Graduate Theses and Dissertations

Examining opportunities for climate resilience requires an evaluation of lived experiences of environmental change which benefits from transdisciplinary perspectives. This research examines those experiences through the lens of anthropogenic change which allows for a critical focus on local conditions of environmental change specific to Placencia Village, Belize, over time. An anthropology of the anthropocene framing shifts the focus from the results of the global phenomenon responsible for climate change to the implications of local human-caused environmental change on natural systems, human-environment relationships and dynamics, and the future trajectories of these relationships. Through the use of interviews, go-alongs, participatory mapping, transdisciplinary …

Digital Humanities: Using Computational Methods On Literature To Understand Human-Water Relations, Ariel Yang

Cybersecurity Undergraduate Research Showcase

By using computational techniques to analyze literature, deeper insights can be gained into human-water relationships across different historical and cultural contexts. Natural Language Processing (NLP) and other data science methods can explore applications of traditional ecological knowledge (TEK) and underlying emotions or beliefs in literature to help understand sustainability. Protecting this sensitive cultural data through ethical applications can further secure future implementations of policies, urban planning, and environmental relationships.

Intelligent Noise Suppression For 3d Post-Stack Seismic Data Of The Junggar Basin, Mao Haibo, Zhou Xin, Li Xiaofeng, Pan Long, Lin Juan, Liu Dawei, Wang Xiaokai

Coal Geology & Exploration

Objective The Junggar Basin is recognized as a significant petroliferous basin in China, and its hydrocarbon exploration targets have shifted to deeper strata. However, the 3D seismic data of this basin suffer from low signal-to-noise ratios (SNRs) and high data volumes due to the basin's complex near-surface conditions, the great depths of exploration targets, and seismic data acquisition methods characterized by wide azimuths, broadbands, and high density. This complicates the identification of hydrocarbon exploration targets, rendering the improvement in the quality of the 3D seismic data by noise suppression vitally important. Methods The progress in the deep learning theory and …

Developing Linguistic Patterns To Mitigate Inherent Human Bias In Offensive Language Detection, Toygar Tanyel, Besher Alkurdi, Serkan Ayvaz

Turkish Journal of Electrical Engineering and Computer Sciences

With the proliferation of social media, there has been a sharp increase in offensive content, particularly targeting vulnerable groups, exacerbating social problems such as hatred, racism, and sexism. Detecting offensive language use is crucial to prevent offensive language from being widely shared on social media. However, the accurate detection of irony, implication, and various forms of hate speech on social media remains a challenge. Natural language-based deep learning models require extensive training with large, comprehensive, and labeled datasets. Unfortunately, manually creating such datasets is both costly and error-prone. Additionally, the presence of human-bias in offensive language datasets is a major …

A Systematic Review Of The Effects Of Ai-Assisted Moderation On Individuals And Groups, Zehui Yu, Lukas Otto, Dennis Assenmacher, Claudia Wagner

Human-Machine Communication

This review paper provides a conceptualization of AI-assisted content moderation with various degrees of autonomy and summarizes experimental evidence for how different levels of automation in content moderation and related losses of autonomy affect individuals and groups. Our results show that current research predominantly focuses on individuallevel effects, necessitating a shift toward understanding the impact on groups. The study highlights gaps in exploring different levels of AI-assisted moderation interventions and misalignments of different conceptualizations that make comparing research results difficult. The discussion underscores the prevailing emphasis on harmful content removal and advocates for investigating more constructive moderation techniques, emphasizing the …

Designing Customized Loss Functions For Training Deep Neural Networks, Ali Pourramezan Fard

Electronic Theses and Dissertations

This dissertation explores the critical role of loss functions in enhancing the predictive performance of deep machine learning models. Loss functions are an integral element of all the ongoing advances we witness daily in this domain. I design custom loss functions and their impacts on various machine learning tasks, particularly in computer vision.

In the first stage of my research, I aim to improve the prediction performance of deep learning models by providing them with more precise feedback associated with task requirements. This led me to create the concept of assistive loss functions. My first proposed loss function, inspired by …

The View From Ventress - 2023-2024 Academic Year, University Of Mississippi, College Of Liberal Arts

Liberal Arts Newsletters

News from the College of Liberal Arts

Vision-Language Integration For Enhanced Locomotion Mode Prediction, Ehsan Ahmadi

LSU Master's Theses

Wearable exoskeletons offer significant potential in enhancing human mobility in industrial environments. However, their adaptability to dynamic, task-intensive settings presents challenges, especially in accurately predicting locomotion modes such as ladder climbing, stair navigation, low-space movement, and obstacle navigation. This research proposes a multimodal framework that integrates visual data and speech commands to improve locomotion mode prediction in unpredictable environments. Multimodal data was collected using smart glasses, capturing both the user’s perspective (field-of-view, FOV) and voice during locomotion tasks. State-of-the-art models—CLIP, ImageBind, and GPT-4o—process these visual and linguistic inputs to predict locomotion activities. The models were evaluated in zero-shot and fine-tuned …

Dynamic Knowledge Elicitation: Leveraging Student Feedback For Improved Language Model Distillation, Reuven Muller

Master's Theses

Large Language Models (LLMs) have significantly advanced the field of natural language processing but remain resource-intensive and impractical for many organizations. Specialist models offer a viable alternative, often developed through Knowledge Distillation (KD) techniques. However, traditional KD methods rely on predefined static datasets to elicit knowledge from the teacher model, failing to dynamically address the weaknesses of the student model during training. This research introduces two novel methods for adaptive knowledge elicitation: Feedback-Driven Question Generation and Agent-Based Targeted Question Generation. These methods iteratively expand the training dataset based on the student model’s performance, leveraging a teacher model to generate targeted …

Artificial Intelligence Foundation Model Risk Identification And Governance Model From Esg Perspective, Jincheng Shi, Guoyu Wang, Yingchun Wang

Bulletin of Chinese Academy of Sciences (Chinese Version)

The application ecology of artificial intelligence foundation model is rapidly expanding. The environment, society, and governance are facing new challenges and opportunities. Exploring the construction of a governance framework for the development risks of foundation model has important theoretical value and practical significance for promoting the healthy and sustainable development of artificial intelligence. Based on the theories of ESG and artificial intelligence governance, this study analyzes the development benefits and typical risks of foundation model from the perspective of ESG and then constructs a risk governance framework and implementation strategies for artificial intelligence foundation models. This study shows that a …

Gmc-2162 Prompt Engineering And Its Effects On Ai And Human Relationships: A Contemporary Approach, Francis Madu, Naga Janaki Madhav Kadiyala, Nivesh Thallapally

C-Day Computing Showcase

A. Background: Prompt engineering refers to the process of designing and refining input prompts for AI models (especially language models like GPT) to improve their outputs. It has become a critical tool in maximizing the performance and utility of AI models in diverse applications, from customer service to content creation. Beyond technical aspects, the interaction between humans and AI is increasingly shaped by the effectiveness of these prompts. B. Motivation: As AI becomes more integrated into daily life, the way humans interact with AI models is profoundly influenced by prompt engineering. Misaligned prompts can lead to misunderstanding, confusion, or unintended …

Gmr-215 Efficient Sentiment Analysis Using Encoder-Only Transformer, Rohan Jonnalagadda, Srinidhi Kandimalla, Siri Yellu

C-Day Computing Showcase

In the era of social media, sentiment analysis has emerged as a vital instrument for comprehending public opinion, especially on sites like LinkedIn and Twitter. Because user-generated content is informal and noisy, traditional sentiment classification techniques like Naive Bayes and Support Vector Machines sometimes find it difficult to capture context, sarcasm, and long-term interdependence. In order to improve sentiment analysis accuracy for social media datasets with a specific focus on sentiments related to corporate layoffs, this study suggests an encoder-only transformer model. Our method successfully captures intricate phrase patterns and contextual subtleties in textual data by leveraging the self-attention mechanism …

Gmr-7179 Improving Alzheimer’S Detection Via Synthetic Data Generation Using Gpt-4 And Multi-Level Embeddings, Venkata Sai Bhargav Mutala, Imaan Shahid

C-Day Computing Showcase

This study leverages large language models (LLMs), particularly GPT-4, to overcome the data limitations often encountered in Alzheimer’s detection. We utilize GPT-4 for data augmentation, generating synthetic speech transcripts to enhance machine learning model training. Our approach combines fine-tuned BERT embeddings with CLAN-derived linguistic features, as well as sentence-level embeddings, to improve classification performance on the ADReSS2020 dataset. BERT and CLAN features capture detailed linguistic variants, while sentence embeddings offer robust semantic representations, collectively enhancing the accuracy and generalization of the models. Among the classifiers tested, the Random Forest model shows the best performance, achieving an accuracy of 88% with …

Gpr-233 Human-Assisted Ai For Detecting Mental Health Indicators In Social Media, Abm Adnan Azmee, Francis E Nweke, Mason V Pederson

C-Day Computing Showcase

Mental health is essential to overall well-being, and mental illness includes conditions that affect a person’s psychological health, causing significant distress and limiting daily functioning. With advancements in technology, social media has become a platform where individuals openly share their emotions and thoughts, offering a unique window into their psychological states. However, traditional machine learning models struggle to interpret social media data's wide range of linguistic nuances. To analyze this data effectively, collaboration with human experts is crucial. This study proposes an innovative human-AI teaming framework that integrates human expertise with artificial intelligence (AI) to address these challenges. Our framework …

Between Copyright And Computer Science: The Law And Ethics Of Generative Ai, Devin R. Desai, Mark Riedl

Northwestern Journal of Technology and Intellectual Property

Copyright and computer science continue to intersect and clash, but they can coexist. The advent of new technologies such as digitization of visual and aural creations, sharing technologies, search engines, social media offerings, and more, challenge copyright-based industries and reopen questions about the reach of copyright law. Breakthroughs in artificial intelligence research, especially Large Language Models that leverage copyrighted material as part of training, are the latest examples of the ongoing tension between copyright and computer science. The exuberance, rush-to-market, and edge problem cases created by a few misguided companies now raises challenges to core legal doctrines and may shift …

Dualrep: Knowledge Graph Completion By Utilizing Dual Representation Of Relational Paths And Tail Node Density Insights, Haji Gul, Feras Al-Obeidat, Adnan Amin, Muhammad Wasim, Fernando Moreira

All Works

Knowledge graphs (KGs) possess a vital role in enhancing the semantic comprehension of extensive datasets across many fields. It facilitate activities like recommendation systems, semantic searching, and intelligent data mining. However, lacking information can sometimes limit the usefulness of knowledge graphs (KGs), as the lack of relationships between entities could severely limit their practical application. Most existing approaches for KG completion primarily concentrate on embedding-based methods or just use relational paths, neglecting the valuable structural information offered by node density. This research presents an approach that effectively combines relational paths and the density features of tail nodes to enhance the …

Multi-Dimensional Visualization Strategies Using Web Scraping Tools: A Word Formation Synthesis Case Study For Russian Verbs Of Sound, John Simmons

Graduate Student Research & Creative Works

No abstract provided.

Exploring Llm Integration And Its Influence On Agent Behavior And Productivity In Insurance Call Centers, Gerardo L. Wibmer Gonzalez

USF Tampa Graduate Theses and Dissertations

This study investigates the theoretical impacts of supportive AI tools, specifically Large Language Models (LLMs), on agent behavior and communication dynamics in call centers. While technological advancements have streamlined operations, limited research addresses the indirect ways these tools influence agent behavior. Using frameworks like context switching—the cognitive shift required when external stimuli prompt attention shifts—and the Hawthorne effect, where perceived observation modifies behavior, we examine how LLMs shape communication patterns. In call centers, this context switching occurs indirectly as agents adapt to AI note-taking features, whereas in industries like Architecture, Engineering, and Construction (AEC), automation tools prompt more immediate changes …

Gray Advice, Keith Porcaro

Duke Law & Technology Review

Debates over economic protectionism or the technology flavor-of-the-month obscure a simple, urgent truth: people are going online to find help that they cannot get from legal and health professionals. They are being let down, by products with festering trust and quality issues, by regulators slow to apply consumer protection standards to harmful offerings, and by professionals loath to acknowledge changes to how help is delivered. The status quo cannot continue. Waves of capital and code are empowering ever more organizations to build digital products that blur the line between self-help and professional advice. For good or ill, “gray advice” is …

Defending Large Language Models Against Jailbreak Attacks Via Layer-Specific Editing, Wei Zhao, Zhe Li, Yige Li, Jun Sun, Jun Sun

Research Collection School Of Computing and Information Systems

Large language models (LLMs) are increasingly being adopted in a wide range of realworld applications. Despite their impressive performance, recent studies have shown that LLMs are vulnerable to deliberately crafted adversarial prompts even when aligned via Reinforcement Learning from Human Feedback or supervised fine-tuning. While existing defense methods focus on either detecting harmful prompts or reducing the likelihood of harmful responses through various means, defending LLMs against jailbreak attacks based on the inner mechanisms of LLMs remains largely unexplored. In this work, we investigate how LLMs respond to harmful prompts and propose a novel defense method termed Layer-specific Editing (LED) …

Mm‑Forecast: A Multimodal Approach To Temporal Event Forecasting With Large Language Models, Haoxuan Li, Zhengmao Yang, Yunshan Ma, Yi Bin, Yang Yang, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

We study an emerging and intriguing problem of multimodal temporal event forecasting with large language models. Compared to using text or graph modalities, the investigation of utilizing images for temporal event forecasting has not been fully explored, especially in the era of large language models (LLMs). To bridge this gap, we are particularly interested in two key questions of: 1) why images will help in temporal event forecasting, and 2) how to integrate images into the LLM-based forecasting framework. To answer these research questions, we propose to identify two essential functions that images play in the scenario of temporal event …

Balancing Visual Context Understanding In Dialogue For Image Retrieval, Zhaohui Wei, Lizi Liao, Xiaoyu Du, Xinguang Xiang

Research Collection School Of Computing and Information Systems

In the realm of dialogue-to-image retrieval, the primary challenge is to fetch images from a pre-compiled database that accurately reflect the intent embedded within the dialogue history. Existing methods often overemphasize inter-modal alignment, neglecting the nuanced nature of conversational context. Dialogue histories are frequently cluttered with redundant information and often lack direct image descriptions, leading to a substantial disconnect between conversational content and visual representation. This study introduces VCU, a novel framework designed to enhance the comprehension of dialogue history and improve cross-modal matching for image retrieval. VCU leverages large language models (LLMs) to perform a two-step extraction process. It …

Ai And Creativity: Effects Of Culture And Task Emotiveness In Human-Ai Collaboration, Choon Ngee Tan

Dissertations and Theses Collection (Open Access)

Creativity is the driving force behind innovation, propelling individuals and societies toward progress by generating novel ideas and groundbreaking solutions. The emergence of generative AI models, exemplified by GPT-3, offers opportunities to enhance human creativity. This paper explores the potential for unprecedented breakthroughs through the synergy between human intuition and AI-driven creativity, providing practical guidance on leveraging AI to amplify creative capacities. Study 1 finds that AI models trained on data from the U.S. and Chinese cultures exhibit cultural norms, values and cognition of those cultures. Study 2 finds that when humans and AI models of the same culture collaborate …

Class Name Guided Out-Of-Scope Intent Classification, Chandan Gautam, Sethupathy Parameswaran, Aditya Kane, Yuan Fang, Savitha Ramasamy, Suresh Sundaram, Sunil Kumar Sahu, Xiaoli Li

Research Collection School Of Computing and Information Systems

The paper introduces Semantics of Class Labelbased Unsupervised Out of Scope Intent Detection (SCOOS), a novel method aimed at enhancing out-of-scope (OOS) intent classification in task-oriented dialogue systems. Unlike prior approaches that rely solely on indomain (ID) data features, SCOOS leverages semantic cues embedded in class labels to improve classification accuracy. The method entails forming a compact feature space centered around the semantics of class labels by minimizing losses between ID features and class names. SCOOS achieves this by creating a compact feature space centered around class label semantics, achieved through minimizing losses between in-domain (ID) features and class names. …

Large Language Models For Software Engineering: A Systematic Literature Review, Xinyi Hou, Yanjie Zhao, Yue Liu, Zhou Yang, Kailong Wang, Li Li, Xiapu Luo, David Lo, John Grundy, Haoyu Wang

Research Collection School Of Computing and Information Systems

Large Language Models (LLMs) have significantly impacted numerous domains, including Software Engineering (SE). Many recent publications have explored LLMs applied to various SE tasks. Nevertheless, a comprehensive understanding of the application, effects, and possible limitations of LLMs on SE is still in its early stages. To bridge this gap, we conducted a Systematic Literature Review (SLR) on LLM4SE, with a particular focus on understanding how LLMs can be exploited to optimize processes and outcomes. We selected and analyzed 395 research articles from January 2017 to January 2024 to answer four key Research Questions (RQs). In RQ1, we categorize different LLMs …

Cirp: Cross‑Item Relational Pre‑Training For Multimodal Product Bundling, Yunshan Ma, Yingzhi He, Wenjun Zhong, Xiang Wang, Roger Zimmermann, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Product bundling has been a prevailing marketing strategy that is beneficial in the online shopping scenario. Effective product bundling methods depend on high-quality item representations capturing both the individual items' semantics and cross-item relations. However, previous item representation learning methods, either feature fusion or graph learning, suffer from inadequate cross-modal alignment and struggle to capture the cross-item relations for cold-start items. Multimodal pre-train models could be the potential solutions given their promising performance on various multimodal downstream tasks. However, the cross-item relations have been under-explored in the current multimodal pre-train models.To bridge this gap, we propose a novel and simple …

Strength Lies In Differences! Improving Strategy Planning For Non-Collaborative Dialogues Via Diversified User Simulation, Tong Zhang, Chen Huang, Yang Deng, Hongru Liang, Jia Liu, Zujie Wen, Wenqiang Lei, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

We investigate non-collaborative dialogue agents, which are expected to engage in strategic conversations with diverse users, for securing a mutual agreement that leans favorably towards the system’s objectives. This poses two main challenges for existing dialogue agents: 1) The inability to integrate user-specific characteristics into the strategic planning, and 2) The difficulty of training strategic planners that can be generalized to diverse users. To address these challenges, we propose TRIP to enhance the capability in tailored strategic planning, incorporating a user-aware strategic planning module and a population-based training paradigm. Through experiments on benchmark non-collaborative dialogue tasks, we demonstrate the effectiveness …

Thoughts To Target: Enhance Planning For Target-Driven Conversation, Zhonghua Zheng, Lizi Liao, Yang Deng, Ee-Peng Lim, Minlie Huang, Liqiang Nie

Research Collection School Of Computing and Information Systems

In conversational AI, large-scale models excel in various tasks but struggle with target-driven conversation planning. Current methods, such as chain-of-thought reasoning and tree-search policy learning techniques, either neglect plan rationality or require extensive human simulation procedures. Addressing this, we propose a novel two-stage framework, named EnPL, to improve the LLMs’ capability in planning conversations towards designated targets, including (1) distilling natural language plans from target-driven conversation corpus and (2) generating new plans with demonstration-guided in-context learning. Specifically, we first propose a filter approach to distill a high-quality plan dataset, ConvPlan1. With the aid of corresponding conversational data and support from …

Don’T Just Say “I Don’T Know”! Self-Aligning Large Language Models For Responding To Unknown Questions With Explanations, Yang Deng, Yong Zhao, Moxin Li, See-Kiong Ng, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Despite the remarkable abilities of Large Language Models (LLMs) to answer questions, they often display a considerable level of overconfidence even when the question does not have a definitive answer. To avoid providing hallucinated answers to these unknown questions, existing studies typically investigate approaches to refusing to answer these questions. In this work, we propose a novel and scalable self-alignment method to utilize the LLM itself to enhance its response-ability to different types of unknown questions, being capable of not only refusing to answer but also providing explanation to the unanswerability of unknown questions. Specifically, the Self-Align method first employ …