Digital Commons Network^™

Rethinking Cognitive Complexity For Unit Tests: Toward A Readability-Aware Metric Grounded In Developer Perception, Wendkûuni C. Ouédraogo, Yinghua Li, Xueqi Dang, Xin Zhou, Anil Koyuncu, Jacques Klein, David Lo, Tegawendé F. Bissyandé

Research Collection School Of Computing and Information Systems

Automatically generated unit tests-from searchbased tools like EvoSuite or LLMs-vary significantly in structure and readability. Yet most evaluations rely on metrics like Cyclomatic Complexity and Cognitive Complexity, designed for functional code rather than test code. Recent studies have shown that SonarSource's Cognitive Complexity metric assigns nearzero scores to LLM-generated tests, yet its behavior on EvoSuitegenerated tests and its applicability to test-specific code structures remain unexplored. We introduce CCTR, a Test-Aware Cognitive Complexity metric tailored for unit tests. CCTR integrates structural and semantic features like assertion density, annotation roles, and test composition patterns-dimensions ignored by traditional complexity models but critical for …

Flow, Immersion, And Presence: Creating Virtual Reality And Engagement In The Era Of Ubiquitous And Intelligent Technologies, Yi Maggie Guo, Fiona Fui-Hoon Nah, Nannan Xi, Marshall Scott Poole

Research Collection School Of Computing and Information Systems

Researchers use the concepts of flow, immersion, and presence to explain the usage of and engagement (e.g., cognitive absorption) with information technology. In this special issue, we showcase four papers on empirical investigations of flow and immersion, their antecedents, and their outcomes. These papers address research questions that range from investigating the antecedents and consequences of immersion in head-mounted displays of virtual reality, designing for the flow experience in extended reality, studying factors influencing user engagement in the metaverse, and identifying adverse effects of work-related flow. We also provide directions and suggestions for future research.

Vibemus: Proactive Agentic System For Music Personalization, Zhiliang Guo, Teng Tu, Yunshan Ma, Xun Yang

Research Collection School Of Computing and Information Systems

Large language models (LLMs) enable diverse forms of AI-assisted creation, yet they often struggle to bridge the preference-articulation gap: users may provide incomplete or vague intentions or lack the vocabulary to specify what they want, yielding outputs misaligned with true preferences. To address this gap and facilitate music creation in a vibe-centric environment, we introduce VibeMus, a proactive agentic system built on open-source components. The system engages in multi-turn dialogue to progressively determine the music’s emotion, genre, lyrics, and other aspects before generation. Simulated evaluations show that proactive clarification improves alignment with users’ intended nuances. Our approach is training-free, leveraging …

Improving Co-Decoding Based Security Hardening Of Code Llms Leveraging Knowledge Distillation, Dong Li, Shanfu Shu, Meng Yan, Zhongxin Liu, Chao Liu, Xiaohong Zhang, David Lo

Research Collection School Of Computing and Information Systems

Large Language Models (LLMs) have been widely adopted by developers in software development. However, the massive pretraining code data is not rigorously filtered, allowing LLMs to learn unsafe coding patterns. Several prior studies have demonstrated that code LLMs tend to generate code with potential vulnerabilities. The widespread adoption of intelligent programming assistants poses a significant threat to the software development process. Existing approaches to mitigating this risk primarily involve constructing secure data that are free of vulnerabilities and then retraining or fine-tuning the models. However, such an effort is resource intensive and requires significant manual supervision. When the model parameters …

Exploring Parameter-Efficient Fine-Tuning Techniques For Code Generation With Large Language Models, Martin Weyssow, Xin Zhou, Kisub Kim, David Lo, Houari A. Sahraoui

Research Collection School Of Computing and Information Systems

Large language models (LLMs) demonstrate impressive capabilities to generate accurate code snippets given natural language intents in a zero-shot manner, i.e., without the need for specific fine-tuning. While prior studies have highlighted the advantages of fine-tuning LLMs, this process incurs high computational costs, making it impractical in resource-scarce environments, particularly for models with billions of parameters. To address these challenges, previous research explored in-context learning (ICL) and retrieval-augmented generation (RAG) as strategies to guide the LLM generative process with task-specific prompt examples. However, ICL and RAG introduce inconveniences, such as the need for designing contextually relevant prompts and the absence …

Managing Rumors On Electronic Interaction Platforms: How Management Responses Affect Investor Reaction, Runyu Wang, Zili Zhang, Keng Siau, Ziqiong Zhang

Research Collection School Of Computing and Information Systems

This study investigates how listed firms respond to investors’ rumor-related inquiries and examines the impact of these responses on investor reactions, as indicated by subsequent daily abnormal stock returns (ARs). Using a unique dataset of question-and-answer (Q&A) interactions from China’s major e-interaction platforms, established by the stock exchanges, our study provides insights into regulated firm-investor communications in a structured Q&A setting. Unlike informal social media channels, these platforms enable official responses from firm representatives, typically board secretaries, under direct regulatory oversight. By analyzing rumor-related Q&A pairs with regression models and several robustness checks, we find that firms can benefit from …

The Moderating Role Of Trait Online Disinhibition In Exacerbating The Online Victim-Bully Cycle, Sandeeshwara K Tennakoon Appuhamillage Kasturiratna, Andree Hartanto

Research Collection School of Social Sciences

The growth of the internet and social media as a communication platform has increased the incidence of cyberbullying victimisation. While there has been increasing research interest on the antecedents of cyberbullying perpetration, little is known on why some individuals who themself experience cyber victimisation are more likely to become cyberbullying perpetrators. Considering the large number of individuals who experience cyberbullying victimisation in the current, highly digitalised world, this trend of victims turning into bullies poses a worrisome possibility and deserves research attention. One promising individual difference that could potentially moderate the relationship between cyberbullying victimisation can lead to cyberbullying perpetration …

The Moderating Role Of Digital Multi-Sided Platform On The Firm Value Of Traditional Value-Chain Firms, Chee Pheng Lim

Dissertations and Theses Collection (Open Access)

This dissertation examines the strategic value and performance implications for traditional value-chain firms (TVCFs) that adopt digital multi-sided platforms (DMSPs) to augment their existing businesses. DMSPs facilitate interactions among multiple user groups through digital intermediaries, generating self-reinforcing network effects that enhance value creation, reduce operational and transaction costs, and enable collective innovation. Despite the prevalence of platform-based firms, 60% of global unicorns and 7 of the world’s 10 most valuable companies operate digital platforms, empirical research on the value-creation mechanisms of DMSPs remains limited. Anchored in the resource-based view and platform theory, this study investigates how DMSP moderates the relationship …

The Regulatory Challenges Raised By The Evolution From Open Banking To Banking-As-A-Service, Nydia Remolina Leon

Research Collection Yong Pung How School Of Law

In May 2024, a company in California suddenly made headlines across the financial press. Synapse Financial Technologies, once hailed as a pioneer in the next wave of financial innovation, filed for bankruptcy. The company had promised to revolutionize banking by providing “Banking-as-a-Service” (BaaS): an invisible layer of technology that allowed apps to plug directly into the financial system.

Policy Evaluation With Nonlinear Trended Outcomes: Covid-19 Vaccination Rates In The United States, Lynn Bergeland Morgan, Peter C. B. Phillips, Donggyu Sul

Research Collection School Of Economics

This paper discusses pitfalls in two way fixed effects (TWFE) regressions when the outcome variables contain nonlinear and possibly stochastic trend components. If a policy change shifts trend paths of outcome variables, TWFE estimation can distort results and invalidate inference, especially in a context of evolving policy decisions. A robust solution is proposed by allowing for dynamic club membership empirically using a relative convergence test procedure. The determinants of respective club memberships are assessed by panel ordered logit regressions. The approach allows for policy evolution and shifts in outcomes according to a convergence cluster framework with transitions over time and …

Optimizing Initial Screening For Colorectal Cancer Detection With Adherence Behavior, Sarah Yini Gao, Yan He, Ruijie Zhang, Zhichao Zheng, Shao Wei Lam Lam, Emile Tan

Research Collection Lee Kong Chian School Of Business

Background: Two-stage screening programs are widely adopted for early colorectal cancer (CRC) detection, where individuals receiving positive outcomes in the first-stage (initial) test are recommended to undergo a second-stage test (colonoscopy) for further diagnosis. Methods: We study the initial test design—the selection of cutoffs for reporting test outcomes—to balance the trade-off between screening effectiveness (i.e., CRC and polyp detection) and efficiency (i.e., colonoscopy costs), incorporating the fact that not all individuals follow up with a colonoscopy after receiving positive outcomes. We integrate the Bayesian persuasion framework with information avoidance to model this problem and apply it to Singapore's CRC screening …

Why Stressed Candidates Make Better Cross-Cultural Hires, Shilpa Madan

Research Collection Lee Kong Chian School Of Business

When interviewing job candidates, you may look for evidence that they stay calm under pressure and interpret signs of stress as red flags. However, our recent research shows that for cross-cultural positions, those red flags might actually indicate your best potential hires. Candidates who exhibit signs of stress may be better suited to navigate the complexities of cross-cultural roles, suggesting a need to reassess traditional interview evaluations.

Neuro-Ins: A Learning-Based One-Shot Node Insertion For Dynamic Routing Problems, Zhiqin Zhang, Jingfeng Yang, Zhiguang Cao, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

The rise in instant delivery services necessitates efficient route planning in last-mile delivery scenarios, where new orders arrive dynamically and need to be integrated into existing routes. In such contexts, complete re-optimization of routes are not permitted, and node insertion to existing route sequences is the only viable option. However, many existing heuristics for node insertion, such as the Cheapest Insertion (CI) method, are myopic and often result in suboptimal solutions retrospectively. This paper presents Neuro-Ins, an initial yet novel attempt at harnessing a learning-based framework to handle the insertion of new orders for the Pickup and Delivery Problem (PDP). …

Shortcuts Everywhere And Nowhere: Exploring Multi-Trigger Backdoor Attacks, Yige Li, Jiabo He, Hanxun Huang, Jun Sun, Xingjun Ma, Yu-Gang Jiang

Research Collection School Of Computing and Information Systems

Backdoor attacks have become a significant threat to the pre-training and deployment of deep neural networks (DNNs). Although numerous methods for detecting and mitigating backdoor attacks have been proposed, most rely on identifying and eliminating the “shortcut” created by the backdoor, which links a specific source class to a target class. However, these approaches can be easily circumvented by designing multiple backdoor triggers that create shortcuts everywhere and therefore nowhere specific. In this study, we explore the concept of Multi-Trigger Backdoor Attacks (MTBAs), where multiple adversaries leverage different types of triggers to poison the same dataset. By proposing and investigating …

In Crypto We Trust: A Cross-National Comparison Of Factors Affecting Trust In Cryptocurrencies, Arif Perdana, W. Eric Lee, Chu Yeong Lim, Gary Pan, Poh Sun Seow

Research Collection School Of Accountancy

Background: Trust in cryptocurrencies is a complex and evolving construct influenced by technological, economic, and social factors. Unlike traditional information systems that rely on institutional trust, cryptocurrency users often depend on underlying technology and community insights. Despite growing interest, existing research has primarily focused on narrow aspects of trust, with limited cross-national perspectives and comprehensive models.

Method: This study investigates how geographic differences impact user trust in cryptocurrencies by surveying 644 respondents from China, Germany, and the USA. Drawing on established trust formation theories, it utilizes a comprehensive model encompassing technological, economic, and social factors.

Results: Our analysis reveals multiple …

How Do Analysts Affect Corporate Innovation? Evidence From Site Visits, Qiang Cheng, Brian Yutao Wang, Holly I. Yang, Zheyuan Zhang

Research Collection School Of Accountancy

While prior studies have examined whether financial analysts affect corporate innovation, there is little research on the mechanism through which this occurs. In this paper, we examine whether and how analysts' questions about innovation during site visits affect corporate innovation. Using a sample of corporate site visits in China, we find that when analysts ask questions about innovation during site visits, firms invest more in R&D in the future. Consistent with knowledge diffusion across firms, this association is stronger when analysts cover more firms in the same industry, when firms share similar technologies as industry peers, and when an innovation-expert …

The Decline And Transformation Of Smallholders In Chinese Agriculture: National Trends, Qian Forrest Zhang, Meiling Wu

Research Collection School of Social Sciences

Using data from official statistical sources and a nationally representative survey, we find that since 2001, China’s agricultural labor force has declined by over 50 percent—a loss of over 200 million smallholders, probably the largest in human history. This trend is the most pronounced in livestock sectors: from 2012 to 2021, 62.5 million smallholder units have disappeared. The 150 million remaining smallholders are mostly commodity producers, participating widely in commodity relations to access agrochemicals, machinery, rental land, and wage labor to produce for markets. The ageing smallholder population also face challenges in intergenerational reproduction, which will accelerate its decline.

Accordance And Conflict Between Religious And Scientific Precautions Against Covid-19 In 27 Societies, T. Samore, D.M.T. Fessler, A.M. Sparks, C. Holbrook, L. Aarøe, C.G. Baeza, M.T. Barbato, P. Barclay, R. Berniūnas, J. Contreras-Garduño, B. Costa-Neves, M.P. Grazioso, P. Elmas, P. Fedor, Li, Norman P., R. Fernández-Morales, L. Garcia-Marques, P. Giraldo-Perez, P. Gul, F. Habacht

Research Collection School of Social Sciences

Meaning-making systems underlie perceptions of the efficacy of threat-mitigating behaviors. Religion and science both offer threat mitigation, yet these meaning-making systems are often considered incompatible. Do such epistemological conflicts swamp the desire to employ diverse precautions against threats? Or do individuals—particularly individuals who are highly reactive to threats—hedge their bets by using multiple threat-mitigating practices despite their potential epistemological incompatibility? Complicating this question, perceptions of conflict between religion and science likely vary across cultures; likewise, pragmatic features of precautions prescribed by some religions make them incompatible with some scientifically-based precautions. The COVID-19 pandemic elicited diverse precautions thus providing an opportunity …

A Multifaceted Nudge-Based Intervention To Reduce Smartphone Use: Findings From A Randomized Cross-Over Trial, K. T. A. Sandeeshwara Kasturiratna, Yi Jing Chua, Andree Hartanto

Research Collection School of Social Sciences

The widespread use of smartphones has raised concerns about problematic usage behaviors, including excessive screentime and frequent phone checking, which can disrupt daily functioning. This study aimed to evaluate the effectiveness of a multifaceted nudge-based intervention in reducing smartphone use through subtle changes to smartphone settings. The intervention incorporated adjustments such as disabling face/touch ID, setting longer passwords, switching to grayscale mode, and removing social media apps from the home screen. A randomized within-subject cross-over trial with counterbalancing was conducted over a two-week period using a daily diary approach paired with objective smartphone usage data to track the impact of …

Prevalence Of Depression, Anxiety, Burden, Burnout, And Stress In Informal Caregivers: An Umbrella Review Of Meta-Analyses, Xun Ci Soh, Andree Hartanto, Naoki Ling, Mayr Reyes, Lester Sim, Nadyanna M. Majeed

Research Collection School of Social Sciences

Informal caregiving is a physically and emotionally taxing role that has a profound impact on caregivers’ emotional and mental well-being. Previous research has examined the mental health of caregivers and found elevated levels of depression, anxiety, burden, burnout, or stress. However, the global rates of the five mental health factors and/or outcomes (i.e., depression, anxiety, burden, burnout, and stress) among caregivers remain largely unclear. Therefore, this umbrella review examines the prevalence of these factors and/or outcomes, including various subgroup factors, such as gender, medical and/or psychological condition of the care recipient, region, and assessment tools. A systematic search was conducted …

Reimagining Academic Assessment In The Age Of Ai, Matthew Hammerton

Research Collection School of Social Sciences

In ‘Reimagining Academic Assessment in the Age of AI’, Matthew Hammerton examines the challenges and opportunities posed by generative AI for higher education assessment. He critiques common responses like banning AI, reverting to in-class exams, or abandoning essays altogether, arguing that they fail to preserve the deeper pedagogical goals of higher order, independent thinking. Instead, Hammerton proposes a guiding principle of intellectual responsibility: students should be accountable for explaining and defending each major choice in their work—regardless of whether they use AI tools. To operationalise this, he advocates for reintegrating oral examinations (‘vivas’) alongside written essays. In this model, students …

What Are The Limits To Smart Innovation? The Digital Underground Initiative And New Frontiers Of “Smartness” In Singapore, Prerona Das, Orlando Woods, Lily Kong

Research Collection College of Integrative Studies

This paper argues that the emergence of new frontiers of smart transformation, such as underground spaces, can pose significant challenges to the implementation of “smart” technological solutions. Exploring the case of Singapore’s Digital Underground initiative, we show that elements such as material complexity, co-ordination between stakeholders, and the availability of appropriate regulatory and administrative frameworks can problematize the implementation of subterranean smart initiatives. By considering multistake-holder perspectives on these “more-than-technological” factors, this paper offers lessons for urban planners, policymakers, and developers in cities around the world that are keen to leverage technology to unlock the potential of underground space.

Public Support For And Reactive Behavior Toward Technological Risk Interventions: An Extension Of Secondary Risk Theory, Hye Kyung Kim, Rui Gu, Fiona Goh, Sonny Rosenthal, Shirley S. Ho

Research Collection College of Integrative Studies

Project Wolbachia is a dengue control initiative in Singapore. Guided by Secondary Risk Theory, this study examined how primary (dengue) and secondary (Wolbachia) threat and coping appraisals shape public support and reactive behavior toward the project. A nationally representative survey of 1,000 Singaporeans showed that support was positively linked to the perceived primary risk, the efficacy of the project and the government, while negatively related to the perceived secondary risk. Reactive behavior was positively associated with perceptions of both primary and secondary risk. Media attention and trust in the government indirectly influenced support and behavior moderated by knowledge.

Optimal Abort Policy For Mission-Critical Systems Under Imperfect Condition Monitoring, Qiuzhuang Sun, Jiawen Hu, Zhi-Sheng Ye

Research Collection College of Integrative Studies

Although most on-demand mission-critical systems are engineered to be reliable to support critical tasks, occasional failures may still occur during missions. To increase system survivability, a common practice is to abort the mission before an imminent failure. We consider optimal mission abort for a system whose deterioration follows a general three-state (normal, defective, failed) semi-Markov chain. The failure is assumed self-revealed, whereas the healthy and defective states have to be inferred from imperfect condition-monitoring data. Because of the non-Markovian process dynamics, optimal mission abort for this partially observable system is an intractable stopping problem. For a tractable solution, we introduce …

Iot Sensors As A Tool For Assessing Spatiotemporal Risk To Extreme Heat, Kijin Seong, Seung Jun Choi, Junfeng Jiao

Research Collection College of Integrative Studies

The safety of urban populations sensitive to extreme heat is under increasing threat. Few studies examine the potential benefits of deploying IoT environmental sensors in the urban context and their integration with large-scale human activity data. This paper examines the deployment of IoT sensors in high-resolution extreme heat risk assessment in the case of Seoul, South Korea. This study conducted spatiotemporal analysis on heat exposure with IoT sensors, compared it with an existing land surface temperature map for validation, combined it with human activity data for risk assessment, and finally discussed the benefits of IoTs in detecting abnormal weather events. …

The Cool, Quiet City Machine Learning Competition: Overview And Results, Clayton Miller, M. Ibrahim, I.S. Akbar, B. Picchetti, Y.X. Chua, M. Frei, F. Biljecki, A. Chong, M. Quintana, C. Fu

Research Collection College of Integrative Studies

The prediction of thermal and noise-based preferences in the urban context is valuable in characterizing interventions to mitigate the challenges of health, productivity, and satisfaction of urban dwellers. The growth of crowd-sourced data and data-driven techniques provides an opportunity to increase the understanding of which machine learning models are most accurate and applicable for this context. This paper outlines the results of a machine learning competition aiming to enhance the accuracy of predicting human comfort in the city context. The competition asks contestants to use contextual data to predict noise distraction and thermal preference in various indoor and outdoor spaces. …

Gti: Graph-Based Tree Index With Logarithm Updates For Nearest Neighbor Search In High-Dimensional Spaces, Ruoyao Ma, Yifan Zhu, Baihua Zheng, Lu Chen, Congcong Ge, Yunjun Gao

Research Collection School Of Computing and Information Systems

Nearest neighbor search (NNS) is fundamental for high-dimensional space retrieval and impacts various fields, such as pattern recognition, information retrieval, recommendation systems, and vector database management. Among existing NNS methods, graph-based methods often excel in query accuracy and efficiency. However, these methods face significant challenges, including high construction costs and difficulties with dynamic data updates. Recent efforts have focused on combining graph methods with hashing, quantization, and tree-based approaches to address these issues, but problems with large index sizes and update performance remain unresolved. In response, this paper proposes GTI, a novel, lightweight, and dynamic graph-based tree index for high-dimensional …

Static Analysis As A Feedback Loop: Enhancing Llm-Generated Code Beyond Correctness, Scott Blyth, Sherlock Licorish, Christoph Treude, Markus Wagner

Research Collection School Of Computing and Information Systems

Large language models (LLMs) have demonstrated impressive capabilities in code generation, achieving high scores on benchmarks such as HumanEval and MBPP. However, these benchmarks primarily assess functional correctness and neglect broader dimensions of code quality, including security, reliability, readability, and maintainability. In this work, we systematically evaluate the ability of LLMs to generate high-quality code across multiple dimensions using the PythonSecurityEval benchmark. We introduce an iterative static analysis-driven prompting algorithm that leverages Bandit and Pylint to identify and resolve code quality issues. Our experiments with GPT-4o show substantial improvements: security issues reduced from >40% to 13%, readability violations from >80% …

Coupling Category Alignment For Graph Domain Adaptation, Nan Yin, Xiao Teng, Zhiguang Cao, Mengzhu Wang

Research Collection School Of Computing and Information Systems

Graph domain adaptation (GDA), which transfers knowledge from a labeled source domain to an unlabeled target graph domain, attracts considerable attention in numerous fields. However, existing methods commonly employ message-passing neural networks (MPNNs) to learn domain-invariant representations by aligning the entire domain distribution, inadvertently neglecting category-level distribution alignment and potentially causing category confusion. To address the problem, we propose an effective framework named Coupling Category Alignment (CoCA) for GDA, which effectively addresses the category alignment issue with theoretical guarantees. CoCA incorporates a graph convolutional network branch and a graph kernel network branch, which explore graph topology in implicit and explicit …

Detecting Defi Fraud With A Graph-Transformer Language Model, Wei Ma, Junjie Shi, Jiaxi Qiu, Cong Wu, Jing Chen, Lingxiao Jiang, Shangqing Liu, Yang Liu, Yang Xiang

Research Collection School Of Computing and Information Systems

With the rapid development of blockchain technology, the widespread adoption of smart contracts—particularly in decentralized finance (DeFi) applications—has introduced significant security challenges, such as reentrancy attacks, phishing, and Sybil attacks. To address these issues, we propose a novel model called TrxGNNBERT, which combines Graph Neural Network (GNN) and the Transformer architecture to effectively handle both graph-structured and textual data. This combination enhances the detection of suspicious transactions and accounts on blockchain platforms like Ethereum. TrxGNNBERT was pre-trained using a masked language model (MLM) on a dataset of 60,000 Ethereum transactions by randomly masking the attributes of nodes and edges, thereby …