Digital Commons Network^™

Seaexam And Seabench: Benchmarking Llms With Local Multilingual Questions In Southeast Asia, Chaoqun Liu, Wenxuan Zhang, Jiahao Ying, Mahani Aljunied, Anh Tuan Luu, Lidong Bing

Research Collection School Of Computing and Information Systems

This study introduces two novel benchmarks, SeaExam and SeaBench, designed to evalu ate the capabilities of Large Language Models (LLMs) in Southeast Asian (SEA) application scenarios. Unlike existing multilingual datasets primarily derived from English translations, these benchmarks are constructed based on real world scenarios from SEA regions. SeaExam draws from regional educational exams to form a comprehensive dataset that encompasses sub jects such as local history and literature. In contrast, SeaBench is crafted around multi turn, open-ended tasks that reflect daily inter actions within SEA communities. Our evalua tions demonstrate that SeaExam and SeaBench more effectively discern LLM performance on …

Towards Explainable And Robust Nlp: Neutrosophic Probability Augmentation In Text Classification, Nabil M. Abdel-Aziz, Mahmoud Ibrahim, Khalid A. Eldrandaly

Neutrosophic Systems with Applications

The rapid growth of textual data necessitates advanced text classification models. However, traditional methods struggle with ambiguity and uncertainty in natural language, reducing classification reliability. To address this, we integrate neutrosophic logic, which explicitly models truth, indeterminacy, and falsity, into a DistilBERT-based text classification framework. Additionally, we employ data augmentation using synonym replacement to enhance generalization. Our approach is evaluated on the AG News dataset, classifying articles into four categories: World, Sports, Business, and Science/Technology. By incorporating neutrosophic attributes, the proposed framework assesses text quality, mitigates uncertainty, and improves robustness against ambiguous inputs. Experimental results demonstrate an accuracy of 94.10%, …

Reverse Modeling In Large Language Models, Sicheng Yu, Yuanchen Xu, Cunxiao Du, Yanying Zhou, Minghui Qiu, Qianru Sun, Hao Zhang, Jiawei Wu

Research Collection School Of Computing and Information Systems

Humans are accustomed to reading and writing in a forward manner, and this natural bias extends to text understanding in auto-regressive large language models (LLMs). This paper investigates whether LLMs, like humans, struggle with reverse modeling, specifically with reversed text inputs. We found that publicly available pre-trained LLMs cannot understand such inputs. However, LLMs trained from scratch with both forward and reverse texts can understand them equally well during inference. Our case study shows that different-content texts result in different losses if input (to LLMs) in different directions---some get lower losses for forward while some for reverse. This leads us …

Hello Again! Llm-Powered Personalized Agent For Long-Term Dialogue, Hao Li, Chenghao Yang, An Zhang, Yang Deng, Xiang Wang, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Open-domain dialogue systems have seen remarkable advancements with the development of large language models (LLMs). Nonetheless, most existing dialogue systems predominantly focus on brief single-session interactions, neglecting the real-world demands for long-term companionship and personalized interactions with chatbots. Crucial to addressing this real-world need are event summary and persona management, which enable reasoning for appropriate long-term dialogue responses. Recent progress in the human-like cognitive and reasoning capabilities of LLMs suggests that LLM-based agents could significantly enhance automated perception, decision-making, and problem-solving. In response to this potential, we introduce a model-agnostic framework, the Long-term Dialogue Agent (LD-Agent), which incorporates three independently …

Worldcuisines: A Massive-Scale Benchmark For Multilingual And Multicultural Visual Question Answering On Global Cuisines, Genta Indra Winata, Et. Al

Research Collection School Of Computing and Information Systems

Vision Language Models (VLMs) often struggle with culture-specific knowledge, particularly in languages other than English and in underrepresented cultural contexts. To evaluate their understanding of such knowledge, we introduce WorldCuisines, a massive-scale benchmark for multilingual and multicultural, visually grounded language understanding. This benchmark includes a visual question answering (VQA) dataset with text-image pairs across 30 languages and dialects, spanning 9 language families and featuring over 1 million data points, making it the largest multicultural VQA benchmark to date. It includes tasks for identifying dish names and their origins. We provide evaluation datasets in two sizes (12k and 60k instances) alongside …

Exploring The Potential Of Large Language Models For Heterophilic Graphs, Yuxia Wu, Shujie Li, Yuan Fang, Chuan Shi

Research Collection School Of Computing and Information Systems

Large language models (LLMs) have presented significant opportunities to enhance various machine learning applications, including graph neural networks (GNNs). By leveraging the vast open-world knowledge within LLMs, we can more effectively interpret and utilize textual data to better characterize heterophilic graphs, where neighboring nodes often have different labels. However, existing approaches for heterophilic graphs overlook the rich textual data associated with nodes, which could unlock deeper insights into their heterophilic contexts. In this work, we explore the potential of LLMs for modeling heterophilic graphs and propose a novel two-stage framework: LLM-enhanced edge discriminator and LLM-guided edge reweighting. In the first …

On Learning Informative Trajectory Embeddings For Imitation, Classification And Regression, Zichang Ge, Changyu Chen, Arunesh Sinha, Pradeep Varakantham

Research Collection School Of Computing and Information Systems

In real-world sequential decision making tasks like autonomousdriving, robotics, and healthcare, learning from observed state-action trajectories is critical for tasks like imitation, classification,and clustering. For example, self-driving cars must replicate humandriving behaviors, while robots and healthcare systems benefitfrom modeling decision sequences, whether or not they come fromexpert data. Existing trajectory encoding methods often focus onspecific tasks or rely on reward signals, limiting their ability togeneralize across domains and tasks.Inspired by the success of embedding models like CLIP andBERT in static domains, we propose a novel method for embeddingstate-action trajectories into a latent space that captures the skillsand competencies in the …

Exploring The Biasing Effects Of Gender On Personality Disorder Diagnoses Formulated By Artificial Intelligence, Zoe Colclough

Student Theses

Gender bias is prevalent in personality disorder assessments, and while artificial intelligence has been posited as a solution to improve diagnostic objectivity and accuracy, the potential for such technologies to propagate human gender bias in mental health contexts remains underexplored. This study investigated the influences of gender bias on the diagnostic performance of ChatGPT-4o for personality disorders using three factorial research designs, which involved experimentally manipulating patient gender in a combined sample of 360 vignettes and case studies. Vignettes were synthesized through a novel artificial intelligence-assisted methodology established for this research, and case studies were identified from the literature. Significant …

The Role Of Individual Values In Bryant University's Sustainability Efforts, John Boccuzzi Iii

Honors Projects in Data Science

This research examines student, faculty, and staff perspectives on sustainability at Bryant University, with the goal of understanding how individual values align with the university's environmental initiatives. The objective is to assess perceptions of Bryant's current sustainability practices, explore how effectively these efforts are communicated across campus, and identify potential gaps between institutional action and community awareness. To achieve this, the study gathers qualitative data through an open-ended survey and applies sentiment analysis to interpret student attitudes toward sustainability. By analyzing these responses alongside Bryant's sustainability marketing efforts, this research will identify gaps between student engagement and institutional messaging The …

A Fuzzy Based Framework For Sustainable Technology Selection In Small-Scale Gold Mining Operations, John M. Kafuku

Tanzania Journal of Engineering and Technology (TJET)

Small-scale gold mining (SSGM) operations in Tanzania has been operating inefficiently due to inadequate mining processing technologies, poor working tools, lack of enough capital, and insufficient electricity. Despite the efforts made by different stakeholders in boosting the sustainability of SSGM yet the sector has not reached the expected goal. This paper proposes a framework for appropriate technology selection to help small scale gold miners in evaluating various gold mineral processing technologies. The framework utilizes the fuzzy logic set theory for technology evaluation and selection. The developed framework for technology selection upon validation provided results that technology adequacy of more than …

Detecting Ideological Bias In Trump's Tweets: Trump's Peace Plan As A Case Study, Amer Qasem

Middle East Journal of Communication Studies

This study aims to explore the extent of ideological bias in the discourse of elected U.S. President Donald Trump on social media platforms, specifically Twitter (now known as X), regarding his peace plan, commonly referred to as the "Deal of the Century." The study employs Critical Discourse Analysis (CDA) to examine all of Trump’s tweets related to the Deal of the Century, seeking to answer the following research questions: How did Trump portray the parties affected by the deal in his tweets? What rhetorical strategies did he employ to present the deal? The analysis spans the period from the announcement …

Fact-Based Counter Narrative Generation To Combat Hate Speech, Brian Wilk, Homaira Huda Shomee, Suman Kalyan Maity, Sourav Medya

Computer Science Faculty Research & Creative Works

Online hatred has become an increasingly pervasive issue, affecting individuals and communities across various digital platforms. To combat hate speech in such platforms, counter narratives (CNs) are regarded as an effective method. In recent years, there has been growing interest in using generative AI tools to construct CNs. However, most of the generative models produce generic responses to hate speech and can hallucinate, reducing their effectiveness. To address the above limitations, we propose a counter narrative generation method that enhances CNs by providing non-aggressive, fact-based narratives with relevant background knowledge from two distinct sources, including a web search module. Furthermore, …

Fictional Failures And Real-World Lessons: Ethical Speculation Through Design Fiction On Emotional Support Conversational Ai, Fayle Kollig, Jessica Pater, Fayika Farhat Nova, Casey Fiesler

Health Services and Informatics Research

Conversational artificial intelligence (CAI), which replicates human-to-human interaction as human-to-machine, is increasingly developed to address insufficient access to healthcare. In this paper, we use design fiction methods to speculate on ethical consequences of CAI that offers emotional support to complement or replace mental healthcare. Through a near-future news article about a fictional, failed CAI, we explore safety and privacy concerns associated with mismatches between what an emotional support CAI is advertised to do, what it technically can do, and how it is likely to be used. We pose the following questions to researchers, regulators, and developers: How might we jointly …

A Hybrid Deep Learning-Based Approach For Visual Field Test Forecasting, Ashkan Abbasi, Sowjanya Gowrisankaran, Wei-Chun Lin, Xubo Song, Bhavna Josephine Antony, Gadi Wollstein, Joel Schuman, Hiroshi Ishikawa

Wills Eye Hospital Papers

OBJECTIVE: Longitudinal assessment of visual field (VF) testing is essential in glaucoma management. Conventional VF forecasting methods require numerous prior tests, while deep learning techniques have shown promising results with fewer tests. This study introduces a hybrid deep learning framework to enhance flexibility and accuracy in VF test forecasting.

DESIGN: A retrospective longitudinal study using deep learning-based VF forecasting models.

SUBJECTS AND CONTROLS: A total of 1750 subjects (healthy and glaucoma patients) with 19 437 Humphrey VF (24-2 Swedish Interactive Threshold Algorithm) tests collected from longitudinal glaucoma cohorts at the University of Pittsburgh and New York University.

METHODS: Three deep …

A Comprehensive Intelligent Traffic Monitoring System Based On A Novel Integration Of Neutrosophic Multi-Criteria Decision-Making Techniques, Mai Mohamed, Amira Salam, Rana Muhammad Zulqarnain, Muhammad Gulistan

Neutrosophic Systems with Applications

With the spread of road accidents and traffic congestion that costs countries and governments a lot of money in addition to the loss of human lives, and since traditional methods of monitoring traffic have not been as effective as desired, attention has been drawn to the search for more effective solutions to the problem of monitoring and regulating traffic. With the spread of technology and the Internet of Things, UAVs have emerged as a promising tool for monitoring traffic, as they can fly for a sufficient period and operate in difficult climatic conditions, in addition to their ability to monitor …

A Performance Comparison Between Two Speech-To-Asl-Gesture-Projection Translation Implementations, Alexandra Kashani Motlagh, Shikha Mehta, Ishfaq Ahmad, Addison Clark, Hansheng Lei

Informatics and Engineering Systems Faculty Publications

Millions of people with hearing disabilities use sign language for communication, creating a communication gap with those who are not fluent in ASL (American Sign Language). This paper aims to introduce an ASL interpreter system using a smart-glasses-based augmented reality system. We begin by introducing and comparing two models that translate spoken language into ASL poses. The first system translates spoken text to ASL Gloss, an intermediate representation, before generating ASL poses. The second system directly translates the text to ASL poses. Our analysis shows that using ASL Gloss as an intermediate step significantly improves the translation speed. We then …

Ontologies For Smart Agriculture: A Path Toward Explainable Ai-A Systematic Literature Review, Rima Grati, Najla Fattouch, Khouloud Boukadi

All Works

Smart agriculture has grown significantly over the last few years, particularly with the integration of advanced technologies (e.g., the Internet of Things, robots, artificial intelligence, etc.), leading to the development of intelligent agricultural systems. However, these systems often lack data integration, interoperability, and semantic explainability. Various approaches have been proposed to address these challenges. This study provides a comprehensive literature review that addresses, on the one hand, the use of semantic resources (e.g., semantic web technologies and ontologies) to tackle data integration and interoperability in smart agriculture systems and, on the other hand, the integration of explainable artificial intelligence into …

Expanding The Horizons Of Nonagonal Neutrosophic Numbers As A Versatile Framework For Decision-Making And Scientific Applications In Covid-19, Muhammad Naveed Jafar, Kainat Muniba, Hamiden Abd El-Wahed Khalifa, Fahd Jarad

Neutrosophic Systems with Applications

In this article, the concept of nonagonal neutrosophic numbers has been introducing in the disjunctive frame of reference. We also proposed the dependency and independency of membership function in regards to nonagonal neutrosophic number. We also introduce a new score function and its computation also formulated in a distinct rational viewpoint. We developed the concept of weighted arithmetic averaging operator and weighted geometric averaging operator for nonagonal neutrosophic numbers. It will open new doors for MCDM and develop the concept with new approaches. Additionally, we analyze the effect of COVID-19 for different ages.

Leveraging Hypersoft Set To Optimize Livestock In The Era Of Unmanned Aerial Vehicles, Alaa Salem, Mona Mohamed, Nebojsa Bacanin, Mohamed Abouhawwash

Neutrosophic Systems with Applications

Due to urbanization and industrialization, rapid global change and the potential loss of arable land, agricultural output must rise in production levels and harvest, distribute, and use resources more efficiently. It is believed that using technology on livestock would help meet the expanding population's demand for food. Internet of Things (IoT) and Unmanned Aerial Vehicle (UAV) integration in conventional farming has transformed operations, providing farmers with greater productivity, improved decision-making, and sustainability. We assume that there are enough UAVs to cover the entire pasture, and our goal is to identify the best UAVs. Accordingly, determining the best type of UAVs …

A Comprehensive Intelligent Traffic Monitoring System Based On A Novel Integration Of Neutrosophic Multi-Criteria Decision-Making Techniques, Mai Mohamed, Amira Salam, Rana Muhammad Zulqarnain, Muhammad Gulistan

Neutrosophic Systems with Applications

With the spread of road accidents and traffic congestion that costs countries and governments a lot of money in addition to the loss of human lives, and since traditional methods of monitoring traffic have not been as effective as desired, attention has been drawn to the search for more effective solutions to the problem of monitoring and regulating traffic. With the spread of technology and the Internet of Things, UAVs have emerged as a promising tool for monitoring traffic, as they can fly for a sufficient period and operate in difficult climatic conditions, in addition to their ability to monitor …

Nash Equilibrium Solutions For Continuous Static Games Under Neutrosophic Environment, M. G. Brikaa

Neutrosophic Systems with Applications

Neutrosophic set theory plays an important role in dealing with the impreciseness and inconsistency in data encountered in solving real life problems. This paper presents a novel approach to solving a new class of continuous static games within a neutrosophic framework. In the proposed methodology, the neutrosophic continuous static games are redefined into two separate crisp problems: the lower problem and the upper problem. The study further establishes the necessary conditions for determining equilibrium strategies in neutrosophic continuous static games. To demonstrate its effectiveness and practical applicability, the proposed method is validated through a numerical example.

Revisiting Bipolar Neutrosophic Graph And Interval-Valued Neutrosophic Graph, Takaaki Fujita

Neutrosophic Systems with Applications

Graph theory explores networks composed of nodes (vertices) and their connections (edges). A graph class is a collection of graphs that share common structural properties, defined by specific rules or constraints. This paper examines various models of uncertain graphs, including Fuzzy, Intuitionistic Fuzzy , Neutrosophic, Turiyam Neutrosophic, and Plithogenic Graphs. In particular, this study focuses on Bipolar Graphs and Interval-valued Graphs, analyzing them within the frameworks of Fuzzy, Neutrosophic, Turiyam Neutrosophic, and Plithogenic Graphs.

Assessing The Sustainable Blockchain-Metaverse-Iot Platform In The Healthcare Industry: An Intelligent Decision Support Model, Ibrahim M. Hezam, Ahmed M. Ali, Ibrahim A. Hameed, Karam Sallam, Mohamed Abdel-Basset

Neutrosophic Systems with Applications

Healthcare services must fulfill patients' desires for secure data sharing and high accessibility. Blockchain technology, through blockchain platforms (BPs), can overcome healthcare challenges. This study develops a decision-making methodology for selecting the best BP, by integrating blockchain with IoT and Metaverse, the proposed approach ensures data integrity, quality, privacy and security, secure data sharing, and interoperability. The decision-making methodology uses the multi-criteria decision-making (MCDM) methodology to handle conflicting criteria. Two MCDM methods are used in this study: CRiteria Importance Through Intercriteria Correlation (CRITIC) for weight computation, and Ranking of Alternatives with Weights of Criterion (RAWEC) for alternative ranking. To deal …

A Novel Neutrosophic Decision-Making Approach For Optimizing Metaverse Headphone Design: Balancing Technical Performance And User Emotional Needs, Mai Mohamed, Amira Salam, Karam Sallam, Bilal Arain

Neutrosophic Systems with Applications

The concept of the metaverse, which combines various technologies to create a wide range of virtual experiences, has gained significant popularity in recent years. To fully engage in these metaverse environments, users rely on access devices such as virtual reality (VR) headsets and smartphones for augmented reality (AR). These devices must be lightweight, compact, and user-friendly to ensure comfort and enhance customer satisfaction. There is a growing focus on innovating and designing products that not only meet technical requirements but also address the emotional needs of users, ultimately improving the overall experience. Selecting the ideal design for Metaverse headphones is …

Meta-Clustering For Specialized Language Models: Enhancing Contextual Adaptation And Mitigating Hallucinations In Diverse Healthcare Environments, Joshit Mohanty, Vaishali Vaishali, Sandeep Kumar Nayak, Sumit Lahiri

Graduate Student Government Association Research Conference

Large Language Models (LLMs) have significantly advanced conversational AI by enabling dialogic information-seeking and task execution across diverse domains. However, their extensive parameters and broad domain scope lead to “data hallucinations.” These shortcomings are particularly evident in dynamic and diverse environments like India’s healthcare sector, where myriad languages, regional practices, and cultural nuances demand specialized, localized expertise rather than one-size-fits-all generalist models. This paper introduces a meta-clustering framework that integrates Distilled Language Models (DLMs) and Small/Specialized Language Models (SLMs) with meta-learning principles to address these limitations. By drawing on evidence from works such as MedHalu and Med-HALT, the framework seeks …

Shaping Mathematics Identity: An Exploratory Study On Specifications Grading In Calculus I At A Hispanic-Serving Institution, Luis M. Fernandez, Kaitlyn Stephens Serbin, Cristina Villalobos, Shaghayegh Azadi Setayesh, Guillermo Garza

School of Mathematical & Statistical Sciences Faculty Publications

Calculus I courses play a pivotal role in shaping students' STEM pathways, making it essential to adopt pedagogies that foster both achievement and mathematics identity development, particularly among underserved groups such as Hispanic students. This study explores the impact of Specifications Grading, an alternative assessment method where students meet specific course learning objectives through multiple attempts, on students’ mathematics identity development. Through a comparative case study of two Calculus I students at a Hispanic-Serving Institution, one enrolled in a specification graded course and the other in a traditionally-graded course, we examine shifts in their self-perceptions of competence, interest, recognition in …

Stereotyping In Language (Technologies): An Examination Of Racial And Gender Stereotypes In Natural Language And Language Models, Messi Lee

McKelvey School of Engineering Graduate Student Theses & Dissertations

This dissertation examines stereotyping across natural language and language technologies through three interconnected studies. The first chapter applies a contemporary model of race relations from social psychology to investigate America's racial framework within American English, revealing how language encodes hierarchical associations between racial/ethnic groups and attributes of superiority and Americanness. The second chapter extends this analysis to Large Language Models (LLMs), finding that these language technologies portray socially subordinate groups as more homogeneous compared to dominant groups. The third chapter investigates stereotyping in Vision Language Models (VLMs), showing that these language technologies generate more uniform representations for women than men …

Exploring Teacher Knowledge In Meeting The Expectations Of Integrated Content Language Development In Mathematics: A Descriptive Case Study, Carrissa Rozelaar-Miller

Doctoral Dissertations and Projects

The purpose of this descriptive case study was to discover how knowledge is applied to meet the expectations of integrated content and language development in mathematics for Title I elementary teachers at an urban school district in the central United States. Halliday’s theory on systemic functional linguistics guided the research as it relates to the function and purpose of language through meaning and social interaction in different genres. The central research question was: How do Title I elementary teachers apply specific integrated content and language development knowledge in mathematics? The participants consisted of 10 teachers with at least three years …

Multi-Agent Translation Team (Matt): Enhancing Low-Resource Language Translation Through Multi-Agent Workflow, Anishka Peter, Mai Dang, Michael Liu, Joaquin Dominguez, Nibhrat Lohia

SMU Data Science Review

Like humans, large language models (LLMs) benefit from revision and refinement, especially for complex tasks requiring critical thinking. Inspired by human collaborative problem-solving, this study introduces a novel multi-agent workflow designed to enhance LLM translations from English to low-resource languages. Multi-Agent Translation Team (MATT) involves the collaboration of agents that are assigned specific roles, such as translator, evaluation coordinator, and various levels of editing, to refine the initial translation into the most desired version possible. The agents work collaboratively in an iterative loop until the translation loss meets a satisfactory threshold. It stands out from other multi-agent workflows by combining …

Climate Change Adaptation Through Economic Democracy, Shriraksha Mohan

Best Integrated Writing

Mohan investigates and combines three different areas of literature and social movements to discuss the problem of climate change and economic adaptation. The value of this work is that it can be used to communicate among three independent types of social movements and inquiries. This can strengthen and connect dispersed yet overlapping efforts in addressing work issues, resource utilization, economic governance and participation, and ecological problems economies face today.