Digital Commons Network^™

Gd3 Synthase Drives Resistance To P53-Induced Apoptosis In Breast Cancer By Modulating Mitochondrial Function, Vivek Anand, Fouad El-Dana, Natalia Baran, Jenny Borgman, Zheng Yin, Stephen T Wong, V Lokesh Battula

Faculty, Staff and Student Publications

TP53 mutations are common in breast cancer (BC) and are associated with poor prognosis. GD3 synthase (GD3S/ST8SIA1), a gene associated with breast cancer stem cells, is upregulated in tumors with p53 mutations. However, the functional relationship between GD3S and p53 is unknown. Here, we show that GD3S levels are highest in breast tumors with specific p53 mutations. Functional studies revealed that wild-type (WT) p53 inhibits GD3S expression, whereas mutation in p53 enhances GD3S expression by upregulating GD3S promoter activity. Moreover, we found that GD3S inhibits wild-type p53-induced apoptosis in BC cells, while BC cells harboring gain-of-function p53 mutations are dependent …

Acquired Resistance In Cancer: Towards Targeted Therapeutic Strategies, Alice Soragni, Erik S Knudsen, Thomas N O'Connor, Cristina E Tognon, Jeffrey W Tyner, Donghwa Kim, Xingxing Zang, Agnieszka K Witkiewicz, David W Goodrich, Dadi Jiang, Seth T Gammon, Christopher D Willey, Paul C Boutros, Vlad C Sandulache, Abdullah A Osman, Jeffrey N Myers, Kamiya Mehla, Pankaj K Singh, Keith S Chan, Hongbo Gao, Himangi Marathe

Faculty, Staff and Student Publications

Development of acquired therapeutic resistance limits the efficacy of cancer treatments and accounts for therapeutic failure in most patients. How resistance arises, varies across cancer types and differs depending on therapeutic modalities is incompletely understood. Novel strategies that address and overcome the various and complex resistance mechanisms necessitate a deep understanding of the underlying dynamics. We are at a crucial time when innovative technologies applied to patient-relevant tumour models have the potential to bridge the gap between fundamental research into mechanisms and timing of acquired resistance and clinical applications that translate these findings into actionable strategies to extend therapy efficacy. …

Addressing Challenges In Developing Treatments For Inherited Retinal Diseases: Recommendations From The Third Monaciano Symposium, Debra A Thompson, K Thiran Jayasundera, Oleg Alekseev, Robin R Ali, Alessia Amato, Isabelle S Audo, James W B Bainbridge, Eyal Banin, Cagri G Besirli, David G Birch, Kari E Branham, Artur V Cideciyan, Stephen P Daiger, Jacque L Duncan, Abigail T Fahim, John G Flannery, Roberto Gattegna, John R Heckenlively, Elise Héon, Alessandro Iannaccone, Naheed W Khan, Samer Khateb, Henry J Klassen, Bart P Leroy, Dario Marangoni, Michel Michaelides, David C Musch, Mark E Pennesi, Simon M Petersen-Jones, Eric A Pierce, Rajesh C Rao, José-Alain Sahel, Paul A Sieving, Enrica Strettoi, Cameron R Strong, Thomas J Wubben, Paul Yang, David N Zacks

Faculty, Staff and Student Publications

Over the past decade, efforts focused on developing genetic therapies for inherited retinal diseases have advanced steadily to clinical trials and the development of a treatment, fueling optimism for the potential of precision medicines to provide safe and effective therapies for these rare conditions. Although several ongoing programs remain poised for success, numerous challenges have negatively impacted the ability to obtain regulatory approvals. The present position paper briefly summarizes recent advances and challenges in developing therapeutics for inherited retinal diseases, and presents a set of recommendations for moving the field forward. The priorities identified are discussed in terms of progress …

White Matter Abnormalities And Cognition In Aging And Alzheimer Disease, Christopher Peter, Aditi Sathe, Niranjana Shashikumar, Kimberly R Pechman, Abigail W Workmeister, Yuankai Huo, Jesse Mez, Logan C Dumitrescu, Katherine A Gifford, Corey J Bolton, Leslie S Gaynor, Shannon L Risacher, Lori L Beason-Held, Yang An, Konstantinos Arfanakis, Guray Erus, Christos Davatzikos, Duygu Tosun-Turgut, Mohamad Habes, Di Wang, Arthur W Toga, Paul M Thompson, Panpan Zhang, Kurt G Schilling, Marilyn Albert, Walter Kukull, Sarah A Biber, Bennett A Landman, Barbara B Bendlin, Sterling C Johnson, Julie Schneider, Lisa L Barnes, David A Bennett, Angela L Jefferson, Susan M Resnick, Andrew J Saykin, Paul K Crane, Michael L Cuccaro, Timothy J Hohman, Derek B Archer, Dimitrios Zaras, Yisu Yang, Alaina Durant, Praitayini Kanakaraj, Michael E Kim, Chenyu Gao, Nancy R Newlin, Karthik Ramadass, Nazirah Mohd Khairi, Zhiyuan Li, Tianyuan Yao, Seo-Eun Choi, Brandon Klinedinst, Michael L Lee, Phoebe Scollard, Emily H Trittschuh, Elizabeth A Sanders

Faculty, Staff and Student Publications

Importance: There has yet to be a large-scale study quantifying the association between white matter microstructure and cognitive performance and decline in aging and Alzheimer disease (AD).

Objective: To investigate the associations between tract-specific white matter microstructure and cognitive performance and decline in aging and AD-related cognitive impairment.

Design setting and participants: This prognostic study of aging and AD, a secondary data analysis of multisite cohort studies, acquired data from 9 cohorts between September 2002 and November 2022. Participants were eligible if they had diffusion-weighted magnetic resonance imaging (dMRI) data, domain-specific cognitive composite z scores, demographic and clinical data, were …

Isolation And Functional Properties Of Highly-Purified N-Terminal Domain Of Human Napi2b By Scalable “Resin Overload” Technique, Daria Savenkova, Irina Makarenko, Daria Nedorezova, Ramziya Kiyamova, Mikhail Bogdanov

Faculty, Staff and Student Publications

Despite the enthusiasm and advances in the purification of native and engineered full-length membrane proteins, little attention has been paid to their fragments which could serve as attractive inspiration for function, regulation, or targeting of full-length membrane protein with therapeutic antibodies (Abs). Production of recombinant fragments of "therapeutic" membrane proteins for early-stage discovery research requires their purification to near homogeneity. It is important not only for the production of biotherapeutic antibodies but also for structural and functional studies of competitive protein-Abs, protein-protein, and lipid-protein interactions which heavily rely on the purity and quality of the isolated protein fragment of interest. …

Single-Cell Proteomic Analysis Reveals Multiple Myeloma Heterogeneity And The Dynamics Of The Tumor Immune Microenvironment In Precursor And Advanced States, Mohamed Kamal, Stephanie N Shishido, Jeremy Mason, Krina Patel, Elisabet E Manasanch, Peter Kuhn

Faculty, Staff and Student Publications

Multiple myeloma (MM) is an aggressive hematologic malignancy arising from plasma cell (PC) proliferation in the bone marrow, progressing from its precursor states MGUS and SMM. Despite therapeutic advances, MM remains incurable, underscoring the need for better risk stratification and early detection. Tumor heterogeneity and dynamic immune microenvironment changes drive progression, yet bulk analyses overlook rare subpopulations critical to disease evolution and resistance. This study employed multiplexed targeted proteomics to characterize bone marrow aspirates (BMA) from 22 patients to observe the change in the distribution of PCs and tumor immune microenvironment (TiME) cells across MM disease states and controls. Bone …

Nature-Inspired Metaheuristics For Optimizing Dose-Finding And Computationally Challenging Clinical Trial Designs, Weng Kee Wong, Yevgen Ryeznik, Oleksandr Sverdlov, Ping-Yang Chen, Xinying Fang, Shouhao Zhou

Faculty, Staff and Student Publications

Metaheuristics are commonly used in computer science and engineering to solve optimization problems, but their potential applications in clinical trial design have remained largely unexplored. This article provides a brief overview of metaheuristics and reviews their limited use in clinical trial settings. We focus on nature-inspired metaheuristics and apply one of its exemplary algorithms, the particle swarm optimization (PSO) algorithm, to find phase I/II designs that jointly consider toxicity and efficacy. As a specific application, we demonstrate the utility of PSO in designing optimal dose-finding studies to estimate the optimal biological dose (OBD) for a continuation-ratio model with four parameters …

Ppp2r1a Mutations Portend Improved Survival After Cancer Immunotherapy, Yibo Dai, Anne Knisely, Mitsutake Yano, Minghao Dang, Emily M Hinchcliff, Annalyn Welp, Matthew Lastrapes, Heng Liu, Zhe Yuan, Chen Wang, Hao Nie, Stephanie Jean, Luis J Montaner, Jiakai Hou, Ami Patel, Shrina Patel, Bryan Fellman, Ying Yuan, Baohua Sun, Renganayaki Krishna Pandurengan, Edwin Roger Parra Cuentas, Joseph Celestino, Yan Liu, Jinsong Liu, R Tyler Hillman, Shannon N Westin, Anil K Sood, Pamela T Soliman, Aaron Shafer, Larissa A Meyer, David M Gershenson, David Vining, Dhakshinamoorthy Ganeshan, Karen Lu, Jennifer A Wargo, Weiyi Peng, Rugang Zhang, Linghua Wang, Amir A Jazaeri

Faculty, Staff and Student Publications

Immune checkpoint blockade (ICB) therapy is effective against many cancers, although resistance remains a major issue and new strategies are needed to improve clinical outcomes1-5. Here we studied ICB response in a cohort of patients with ovarian clear cell carcinoma-a cancer type that poses considerable clinical challenges and lacks effective therapies6-8. We observed significantly prolonged overall survival and progression-free survival in patients with tumours with PPP2R1A mutations. Importantly, our findings were validated in additional ICB-treated patient cohorts across multiple cancer types. Translational analyses from tumour biopsies demonstrated enhanced IFNγ signalling, and the presence of tertiary lymphoid structures at the baseline, …

Celebrating The Fifth Edition Of The International Symposium On Fungal Stress – Isfus, A Decade After Its 2014 Debut, Alene Alder-Rangel, Amanda E A Rangel, Arturo Casadevall, Asiya Gusa, Chaoyang Xue, Chris Todd Hittinger, Consuelo Olivares-Yañez, Deborah Bell-Pedersen, Erica J Washington, Gerhard Braus, Guilhem Janbon, István Pócsi, Jason E Stajich, Jay C Dunlap, Joan W Bennett, Joseph Heitman, Ling Lu, Lucia Landi, Mari L Shinohara, Maurizio Del Poeta, Mavis A Acheampong, Mia R Maltz, Michael C Lorenz, Minou Nowrousian, N Louise Glass, Nichole A Broderick, Nicolás Pedrini, Nir Osherov, R Blake Billmyre, Sabrina Sarrocco, Salomé Leibundgut-Landmann, Vânia Aparecida Vicente, Xiaorong Lin, Xin-Qing Zhao, Yong-Sun Bahn, Zachary A Lewis, Drauzio E N Rangel

Faculty, Staff and Student Publications

The Fifth International Symposium on Fungal Stress (ISFUS) brought together in Brazil many of the leaders in the field of fungal stress responses, from fourteen countries, for four days of outstanding science ranging from basic research to studies with agricultural, medical, industrial, and environmental significance. In addition to the excellent oral and poster presentations, the Symposium organisers ensured that all participants had ample opportunity to engage, socialise, and network to exchange ideas and share research. The conference was enhanced by the world-class venue near Iguazu Falls, probably the greatest natural phenomenon in South America.

Counteracting Cascades Challenge The Heterogeneity-Stability Relationship, Jordi Sola, Tom P. Fairchild, Matthew J. Perkins, James C. Bull, John N. Griffin

School of Biological and Marine Sciences

Spatial environmental heterogeneity is widely assumed to enhance ecological stability by promoting refugia, biodiversity and asynchrony. Yet, we lack field experiments testing this fundamental relationship and its underlying mechanisms in naturally assembled multitrophic systems. To address this gap, we monitored experimental substrates replicating topographic heterogeneity on a rocky shore over 3 years. Contrary to theory, heterogeneity showed no net effect on community stability due to four counteracting pathways. Heterogeneity increased stability by (i) providing refugia that enhanced population stability and (ii) boosting species richness, which promoted asynchrony. At the same time, it decreased stability by (iii) reducing a dominant non-native …

The Microrna Mir-30a Blocks Adipose Tissue Fibrosis Accumulation In Obesity, Michael J Bolt, Sung Yun Jung, Kang Ho Kim, Kai Sun, Huaizhu Wu, Samuel Klein, Sean M Hartig

Faculty, Staff and Student Publications

White adipose tissue (WAT) fibrosis occurring in obesity contributes to the inflammatory and metabolic comorbidities of insulin resistance and type 2 diabetes, yet the mechanisms involved remain poorly understood. Here, we report a role for the broadly conserved miRNA miR-30a as a regulator of WAT fibrosis and systemic glucose metabolism. Mice modified to express miR-30a at elevated levels in adipose tissues maintain insulin sensitivity coupled with reduced fatty liver disease when fed a high-fat diet. These effects were attributable to cell-autonomous functions of miR-30a that potently increase expression of adipocyte-specific genes. Proteomic screening revealed miR-30a limits profibrotic programs in subcutaneous …

Womanhood, Survival, And The American Revolution: The Correspondence Between John And Abigail Adams And Henry And Lucy Knox, Kristin Jaques

All Graduate Theses and Dissertations, Fall 2023 to Present

The letters of Lucy and Henry Knox and Abigail and John Adams sent between 1775 and 1782 show how women stepped out of their expected gendered roles with thoughts of survival, not because they wanted to challenge the patriarchal society prevalent in the American colonies during the American Revolution.

Microplastics And Other Plastic-Like Microdebris In Sediments From The Dart River And Estuary, Southwest England, Maia Killian, Andrew Turner

School of Geography, Earth and Environmental Sciences

Surficial sediment samples (n = 10) collected from a transect of a river-estuary (Dart) in southwest England have been analysed for geochemical characteristics, anthropogenic metals (Cr, Cu, Sn, Pb, Zn), and microplastics (∼50 μm–5 mm) and other microplastic-like debris. Results indicated the finest grains towards the lower river and upper estuary, a seaward increase in sedimentary Ca, and loss on ignition ranging from about 0.7 to 4 %. Anthropogenic metals were heterogeneously distributed throughout the catchment with maximum concentrations in the river or lower estuary. Microplastics were dominated by fibres, with concentrations ranging from 110 to 1800 kg−1 and composition …

Curved Confinement Directs Anchoring-Mediated Structural Transitions In Highly Chiral Liquid Crystal Shells, Sepideh Norouzi, Jeremy Money, Stiven Villada-Gil, José Martínez-Gonález, Monirosadat Sadati

Faculty Publications

Cholesteric liquid crystals (CLCs) confined in curved geometries exhibit a rich spectrum of defectmediated morphologies governed by the interplay between chirality, curvature, surface anchoring, and confinement. This study systematically investigates structural transitions in highly chiral CLC shells under asymmetric anchoring conditions, focusing on the effects of shell thickness and curvature on pitch axis reorientation and defect formation. Utilizing microfluidic techniques, we generate core–shell droplets with independently tunable anchoring at inner and outer aqueous interfaces. Transitioning from planar–planar to planar-homeotropic boundary conditions via surfactant-mediated modulation induces profound reorganizations in the director field, giving rise to focal conic domains (FCDs), stripe patterns, …

A Sociological Examination Of Masculinity In Korean Boys' Love Dramas, Jeremiah E. Magoncia

Korean Studies Department Faculty Publications

Korean Boys' Love (K-BL) dramas represent a growing subcategory within South Korean media, exploring romantic relationships between male characters and challenging historically rigid gender norms shaped by Confucian values, modernity, and hegemonic masculinity. Grounded in Social Constructionism and Queer Theory, this chapter employs content analysis on five K-BL dramas—Wish You, To My Star, Semantic Error, Jun and Jun, and Love for Love's Sake—to analyze how masculinity is constructed, portrayed, and subverted. The study reveals that K-BL dramas depict a spectrum of "Seme" and "Uke" roles that soften rigid gender binaries through emotional openness, gentleness, …

Dealuminated Metakaolin In Supplementary Cementitious Material And Alkali-Activated Systems: A Review, Amany G. Micheal Prof., Mostafa Adel Elsebaei Eng., Maria Mavroulidou Prof., Maria Centeno Assoc. Prof., Rabee Shamass Senior Lec., Ottavia Rispolio Mrs.

Architectural Engineering

This paper presents a comprehensive review of dealuminated metakaolin (DK), a hazardous industrial by-product generated by the aluminium sulphate (alum) industry and evaluates its potential as a component in cementitious systems for the partial or full replacement of Portland cement (PC). Positioned within the context of waste valorisation in concrete, the review aims to establish a critical understanding of DK formation, properties, and reactivity, particularly its pozzolanic potential, to assess its suitability for use as a supplementary cementitious material (SCM), or as a precursor in alkali-activated cement (AAC) systems for concrete. A systematic methodology is used to extract and synthesise …

Evaluation Of The Impact Of Project-Based Vocational Learning Programs At Vocational Schools, Muhammad Sadam Rizkylillah, Soeprijanto Soeprijanto, Muksin Muksin

Jurnal Pendidikan: Teori, Penelitian, dan Pengembangan

This study evaluates the impact (effectiveness and outcome) of project-based vocational learning program in Electrical Power Installation programs (TITL) at vocational schools in Tangerang, Indonesia. Using the Goal Free Evaluation model and a mixed-methods approach, quantitatively, the study found that 93,93% of the program ran well. It identified 33 positive impacts that matched the program's goals, 2 negative impacts, and 6 unexpected positive outcomes for students such as curiosity, confidence, safety awareness, entrepreneurship, responsibility, and honesty. Addressing high vocational graduate unemployment, this research fills a gap by focusing on real impacts rather than predefined goals. The program is recommended to …

Dreamanime: Learning Style-Identity Textual Disentanglement For Anime And Beyond, Chenshu Xu, Yangyang Xu, Huaidong Zhang, Xuemiao Xu, Shengfeng He

Research Collection School Of Computing and Information Systems

Text-to-image generation models have significantly broadened the horizons of creative expression through the power of natural language. However, navigating these models to generate unique concepts, alter their appearance, or reimagine them in unfamiliar roles presents an intricate challenge. For instance, how can we exploit language-guided models to transpose an anime character into a different art style, or envision a beloved character in a radically different setting or role? This paper unveils a novel approach named DreamAnime, designed to provide this level of creative freedom. Using a minimal set of 2-3 images of a user-specified concept such as an anime character …

Solving Two-Stage Stochastic Integer Programs Via Representation Learning, Yaoxin Wu, Zhiguang Cao, Wen Song, Yingqian Zhang

Research Collection School Of Computing and Information Systems

Solving stochastic integer programs (SIPs) is extremely intractable due to the high computational complexity. To solve two-stage SIPs efficiently, we propose a conditional variational autoencoder (CVAE) for scenario representation learning. A graph convolutional network (GCN) based VAE embeds scenarios into a low-dimensional latent space, conditioned on the deterministic context of each instance. With the latent representations of stochastic scenarios, we perform two auxiliary tasks: objective prediction and scenario contrast, which predict scenario objective values and the similarities between them, respectively. These tasks further integrate objective information into the representations through gradient backpropagation. Experiments show that the learned scenario representations can …

Preface, Special Issue For The 16th International Conference On Graph Transformation (Icgt 2023), Maribel Fernández, Christopher M. Poskitt

Research Collection School Of Computing and Information Systems

This special issue contains six extended versions of papers presented at the 16th International Conference on Graph Transformation (ICGT 2023), held in Leicester, UK, on 19–20 July 2023. The conference was part of STAF 2023 (Software Technologies: Applications and Foundations) and was held under the auspices of the European Association for Theoretical Computer Science (EATCS), the European Association of Software Science and Technology (EASST), and the IFIP Working Group 1.3, Foundations of Systems Specification.

Evaluating And Mitigating Linguistic Discrimination In Large Language Models: Perspectives On Safety Equity And Knowledge Equity, Guoliang Dong, Haoyu Wang, Jun Sun, Xinyu Wang

Research Collection School Of Computing and Information Systems

By training on text in various languages, large language models (LLMs) typically possess multilingual support and demonstrate remarkable capabilities in solving tasks described in different languages. However, LLMs can exhibit linguistic discrimination due to the uneven distribution of training data across languages. That is, LLMs are hard to keep the consistency of responses when faced with the same task but depicted in different languages. In this study, we first explore the consistency in the LLMs’ outputs responding to queries in various languages from two aspects: safety and quality. We conduct this analysis with two datasets (AdvBench and NQ) based on …

Cami: A Counselor Agent Supporting Motivational Interviewing Through State Inference And Topic Exploration, Yizhe Yang, Palakorn Achananuparp, Heyan Huang, Jing Jiang, Phey Ling Kit, Nicholas Gabriel Lim, Cameron Shi Ern Tan, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Conversational counselor agents have become essential tools for addressing the rising demand for scalable and accessible mental health support. This paper introduces CAMI, a novel automated counselor agent grounded in Motivational Interviewing (MI) – a client-centered counseling approach designed to address ambivalence and facilitate behavior change. CAMI employs a novel STAR framework, consisting of client’s state inference, motivation topic exploration, and response generation modules, leveraging large language models (LLMs). These components work together to evoke change talk, aligning with MI principles and improving counseling outcomes for diverse clients. We evaluate CAMI’s performance through both automated and expert evaluations, utilizing simulated …

Towards Context-Aware Traffic Classification Via Time-Wavelet Fusion Network, Ziming Zhao, Zhuoxue Song, Xiaofei Xie, Zhaoxuan Li, Jiongchi Yu, Fan Terry Zhang, Tingting Li

Research Collection School Of Computing and Information Systems

Encrypted traffic classification occupies a significant role in cybersecurity and network management. The existing encrypted traffic classification technology mostly relies on intra-flow semantics for extracting features. However, considering that some attack behaviors inherently have similar patterns to legitimate behaviors, and powerful adversaries could simulate benign users to conceal their attack intentions, intra-flow features may be similar between different categories. In this paper, we propose TrafficScope, a time-wavelet fusion network based on Transformer to enhance the performance of encrypted traffic classification. Specifically, in addition to using intra-flow semantics, TrafficScope also extracts contextual information to construct more comprehensive representations. Moreover, to cope …

Assessing The Robustness Of Test Selection Methods For Deep Neural Networks, Qiang Hu, Yuejun Guo, Xiaofei Xie, Maxime Cordy, Wei Ma, Mike Papadakis, Lei Ma, Yves Le Traon

Research Collection School Of Computing and Information Systems

Regularly testing deep learning-powered systems on newly collected data is critical to ensure their reliability, robustness, and efficacy in real-world applications. This process is demanding due to the significant time and human effort required for labeling new data. While test selection methods alleviate manual labor by labeling and evaluating only a subset of data while meeting testing criteria, we observe that such methods with reported promising results are simply evaluated, e.g., testing on original test data. The question arises: are they always reliable? In this article, we explore when and to what extent test selection methods fail. First, we identify …

How To Enable Effective Cooperation Between Humans And Nlp Models: A Survey Of Principles, Formalizations, And Beyond, Chen Huang, Yang Deng, Wenqiang Lei, Jiancheng Lv, Tat-Seng Chua, Jimmy Huang

Research Collection School Of Computing and Information Systems

With the advancement of large language models (LLMs), intelligent models have evolved from mere tools to autonomous agents with their own goals and strategies for cooperating with humans. This evolution has birthed a novel paradigm in NLP, i.e., human-model cooperation, that has yielded remarkable progress in numerous NLP tasks in recent years. In this paper, we take the first step to present a thorough review of human-model cooperation, exploring its principles, formalizations, and open challenges. In particular, we introduce a new taxonomy that provides a unified perspective to summarize existing approaches. Also, we discuss potential frontier areas and their corresponding …

Fact-Audit: An Adaptive Multi-Agent Framework For Dynamic Fact-Checking Evaluation Of Large Language Models, Hongzhan Lin, Yang Deng, Yuxuan Gu, Wenxuan Zhang, Jing Ma, See-Kiong Ng, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Large Language Models (LLMs) have significantly advanced the fact-checking studies. However, existing automated fact-checking evaluation methods rely on static datasets and classification metrics, which fail to automatically evaluate the justification production and uncover the nuanced limitations of LLMs in fact-checking. In this work, we introduce FACT-AUDIT, an agent-driven framework that adaptively and dynamically assesses LLMs’ fact-checking capabilities. Leveraging importance sampling principles and multi-agent collaboration, FACT-AUDIT generates adaptive and scalable datasets, performs iterative model-centric evaluations, and updates assessments based on model-specific responses. By incorporating justification production alongside verdict prediction, this framework provides a comprehensive and evolving audit of LLMs’ factual reasoning …

Mpo: Multilingual Safety Alignment Via Reward Gap Optimization, Weixiang Zhao, Yulin Hu, Yang Deng, Tongtong Wu, Wenxuan Zhang, Jiahe Guo, An Zhang, Yanyan Zhao, Bing Qin, Tat-Seng Chua, Ting Liu

Research Collection School Of Computing and Information Systems

Large language models (LLMs) have become increasingly central to AI applications worldwide, necessitating robust multilingual safety alignment to ensure secure deployment across diverse linguistic contexts. Existing preference learning methods for safety alignment, such as RLHF and DPO, are primarily monolingual and struggle with noisy multilingual data. To address these limitations, we introduce Multilingual reward gaP Optimization (MPO), a novel approach that leverages the well-aligned safety capabilities of the dominant language (e.g., English) to improve safety alignment across multiple languages. MPO directly minimizes the reward gap difference between the dominant language and target languages, effectively transferring safety capabilities while preserving the …

A Comprehensive Analysis Of Evolving Permission Usage In Android Apps: Trends, Threats, And Ecosystem Insights, Ali Alkinoon, Trung Cuong Dang, Ahod Alghuried, Abdulaziz Alghamdi, Soohyeon Choi, Manar Mohaisen, An Wang, Saeed Salem, David Mohaisen

Research Collection School Of Computing and Information Systems

The proper use of Android app permissions is crucial to the success and security of these apps. Users must agree to permission requests when installing or running their apps. Despite official Android platform documentation on proper permission usage, there are still many cases of permission abuse. This study provides a comprehensive analysis of the Android permission landscape, highlighting trends and patterns in permission requests across various applications from the Google Play Store. By distinguishing between benign and malicious applications, we uncover developers’ evolving strategies, with malicious apps increasingly requesting fewer permissions to evade detection, while benign apps request more to …

Explainable Multimodal Sentiment Analysis Of Social Media Visual Content For Child Safety, Yee Sen Tan, Zhaoxia Wang

Research Collection School Of Computing and Information Systems

Ensuring the safety and well-being of children is increasingly important, especially in a world where visual content is pervasive. This paper proposes a novel multimodal, multilingual, and multiclass sentiment analysis method for social media content, aimed at improving content moderation for child safety. Our approach integrates textual, visual, and audio data from videos, categorizing sentiment into four levels: positive, slightly negative, negative, and strongly negative, enabling granular detection of harmful content. To enhance explainability and trust, we also leverage interpretable mechanisms to analyze the contributions of each modality. Evaluation of our method demonstrates strong generalization across diverse video types, and …

Information-Theoretic Detection Of Unusual Source Code Changes, Adriano Torres, Markus Wagner, Christoph Treude, Sebastian Baltes

Research Collection School Of Computing and Information Systems

The code base of software projects evolves essentially through inserting and removing information to and from the source code. We can measure this evolution via the elements of information—tokens, words, nodes—of the respective representation of the code. In this work, we approach the measurement of the information content of the source code of open-source projects from an information-theoretic standpoint. Our focus is on the entropy of two fundamental representations of code: tokens and abstract syntax tree nodes, from which we derive definitions of textual and structural entropy. We proceed with an empirical assessment where we evaluate the evolution patterns of …