Digital Commons Network^™

Attackg+: Boosting Attack Graph Construction With Large Language Models, Yongheng Zhang, Tingwen Du, Yunshan Ma, Xiang Wang, Yi Xie, Guozheng Yang, Yuliang Lu, Ee‑Chien Chang

Research Collection School Of Computing and Information Systems

Attack graph construction seeks to convert textual cyber threat intelligence (CTI) reports into structuredrepresentations, portraying the evolutionary traces of cyber attacks. Even though previous research hasproposed various methods to construct attack graphs, they generally suffer from limited generalizationcapability to diverse knowledge types as well as requirement of expertise in model design and tuning.Addressing these limitations, we seek to utilize Large Language Models (LLMs), which have achieved enormoussuccess in a broad range of tasks given exceptional capabilities in both language understanding and zeroshot task fulfillment. Thus, we propose a fully automatic LLM-based framework to construct attack graphsnamed: AttacKG+. Our framework consists …

Enhanced Sample Selection With Confidence Tracking: Identifying Correctly Labeled Yet Hard-To-Learn Samples In Noisy Data, Weiran Pan, Wei Wei, Feida Zhu, Yong Deng

Research Collection School Of Computing and Information Systems

We propose a novel sample selection method for image classification in the presence of noisy labels. Existing methods typically consider small-loss samples as correctly labeled. However, some correctly labeled samples are inherently difficult for the model to learn and can exhibit high loss similar to mislabeled samples in the early stages of training. Consequently, setting a threshold on per-sample loss to select correct labels results in a trade-off between precision and recall in sample selection: a lower threshold may miss many correctly labeled hard-to-learn samples (low recall), while a higher threshold may include many mislabeled samples (low precision). To address …

Multicultural Experience And Young Chinese People’S Subjective Well-Being: An Indirect Effect Through Self-Construal, Xiaoyu Dai, Davelle Jing Yi Lee, Chi-Ying Cheng

Research Collection School of Social Sciences

Traditional Chinese culture promotes ingroup harmony and personal conformity, which facilitates a cultural tendency of perceived self-other similarity among Chinese people’s self-concepts. However, with increased exposure to foreign cultures due to globalization, many young Chinese may see themselves as more unique in relation to others via contrast and comparison. This shift in self-construal may facilitate the pursuit of a more positive self-view and enhanced personal happiness. Thus, it is expected that among Chinese people, multicultural experience would be positively linked to construing the self as different from others, which, in turn, would predict higher self-esteem and subjective well-being. An online …

How Should Universities Serve The 100-Year Life Span?, Lily Kong, Hamish Coates

Research Collection School of Social Sciences

Based on the 2024 lectures delivered by SMU President Professor Lily Kong as part of the 15th Singapore Institute for Policy Studies-Nathan Lecture Series titled ‘Universities Reinvented: Shaping Legacy and Impact for a New World’, the article highlights the need for universities to find a foothold in the next 100 years by moving beyond issuing credentialing to offering lifelong enabling experiences. It advocates for a shift towards holistic education that emphasises independence, humanities, interdisciplinary depth, and policy innovation. Additionally, it calls for government policy reform, stakeholder engagement, and a clear break from the global rankings race to ensure universities stay …

Exploring Business' And Consumers' Engagement On E-Commerce Platforms, Yichen Wang

Dissertations and Theses Collection (Open Access)

The rapid growth of e-commerce platforms has changed how consumers make decisions and how sellers advertise their products. This dissertation examines two key aspects: consumer choice modeling in dynamic online shopping environments and sellers' strategic advertising decisions in competitive markets.

The first essay introduces a Hierarchical Bayesian model to improve the understanding of Individual-Variant Choice Sets (IVCS), where consumers' choice sets change over time and include multiple selections. Traditional models struggle to account for these complexities due to restrictive assumptions. The proposed model introduces a flexible covariance structure that captures perceived similarities between alternatives, leading to better predictive accuracy. Validation …

On/ In The Ground With Vietnamese Communists: The Cu Chi Tunnels In The Classroom, Wen-Qing Ngoei

Research Collection College of Integrative Studies

Undergraduates often arrive in my classes convinced that great men, and a few exceptional women, are the ones who make history. Here, they think of power wielded by presidents, generals, and self-styled geniuses of strategy and diplomacy. (Henry Kissinger’s name invariably bubbles up in class discussions, like it or not.) Perhaps this happens elsewhere? In Singapore, this mindset likely takes root because the country’s record of benevolent authoritarianism underpins the legitimacy and influence of political elites. Equally, male citizens of Singapore perform compulsory military service before entering university and internalize, readily or grudgingly, that rank confers power—it is power—and that …

Attention To Market Expansion: Aspirational Antecedent And Risk-Taking Consequence, Renfei Gao, Man Zhang, Xiaogang He, Zhengyu Li

Research Collection Lee Kong Chian School Of Business

The behavioral theory of the firm (BTOF) emphasizes the goal-directed logic in explaining firms’ risk-taking.Scholars in this tradition typically conceptualize how performance feedback indicates problems and triggers risk-taking behaviors. Yet, the theory remains underdeveloped in understanding decision-makers’ attention to forward-looking goals, such as market expansion, and its impact on firm risk-taking. In this study, we theorize performance feedback as an antecedent to decision-makers’ attention to forward-looking goals, and firm risktaking as a consequence of such attention. Focusing on market expansion as a critical forward-looking goal, we first theorize how performance feedback influences decision-makers’ attention to market expansion. Then, we examine …

From Low Power To Action: Reappraising Powerlessness As An Opportunity Restores Agency, Tianyu He, Michael Schaerer, Trevor A. Foulk, Elizabeth B. Wolf, Winnie Y. Jiang

Research Collection Lee Kong Chian School Of Business

Agentic behaviors are a critical pathway to power in contemporary organizations. Paradoxically, employees who lack power are the least likely to think and act agentically—creating a self-perpetuating cycle of disadvantage. Existing research on facilitating employee agentic behaviors relies on structural solutions that are often out of reach for individual employees. Yet, anecdotal evidence suggests that this view may be incomplete, as some individuals seem to be able to overcome the challenges powerlessness poses without relying on external resources, control, or organizational change. Integrating research on powerlessness and cognitive reappraisal, the present research proposes that cognitively reappraising powerless situations as opportunities …

Csr As Hedging Against Institutional Transition Risk: Corporate Philanthropy After The Sunflower Movement In Taiwan, Yishu Cai, Lori Yue, Fangwen Lin, Shipeng Yan, Haibin Yang

Research Collection Lee Kong Chian School Of Business

Firms with political connections to a regime with an authoritarian history face a dilemma when the regime undergoes a democratic transition. Such connections provide an essential competitive advantage when the regime is in power but become a liability when an institutional transition brings democratic change. This study theorizes that when mass protests expose a regime’s distorted policies favoring elites over others and signal a high probability of regime turnover, firms may hedge against the risks associated with their political connections by engaging in philanthropy. We further contend that this effect is stronger for firms located in regions characterized by the …

Sell-Side Analysts’ Rating Systems, Ohad Kadan, Leonardo Madureira, Rong Wang, Tzachi Zach

Research Collection Lee Kong Chian School Of Business

Different brokers employ a variety of stock recommendations rating systems. There are five- and three-tier systems, as well as different mappings from the economic meaning of recommendations to numerical values in popular data sets such as IBES. We show that the rating systems and their mappings have important implications on the distribution of stock recommendations, investors’ reactions to stock recommendations, the consensus and dispersion of stock recommendations, and measures of analysts’ herding and boldness. We demonstrate how one can control for the variation in rating systems in order to separate the true economic impact of recommendations from the mechanical, non-economic, …

Amazon Versus Walmart: Clash Or Strategic Complementarity, Nirmalya Kumar

Research Collection Lee Kong Chian School Of Business

Nirmalya Kumar explains why Amazon and Walmart will continue to coexist, with Amazon dominating online retail and Walmart shifting to an omnichannel strategy.

Varium: Variational Autoencoder For Multi-Interest Representation With Inter-User Memory, Nhu Thuat Tran, Hady W. Lauw

Research Collection School Of Computing and Information Systems

Frameworks for discovering multiple user interest factors based on Variational AutoEncoder (VAE) has demonstrated competitive recommendation performance. However, as VAE only considers one user as input at a time, sharing across like-minded users may not be adequately facilitated. Moreover, interest sharing between users is not always available and thus, poses a challenge for VAE to explicitly model this information. To resolve this, we introduce an inter-user memory-based mechanism to unsupervisedly discover latent interest sharing between users under VAE framework. Concretely, we design a memory including an array of prototypes, each hypothetically representing a group of users sharing a particular interest. …

Marginal Benefit Driven Rl Teacher For Unsupervised Environment Design, Dexun Li, Wenjun Li, Pradeep Varakantham

Research Collection School Of Computing and Information Systems

Training generally capable agents in complex environments is a challenging task that involves identifying the “right” environments at the training stage. Recent research has highlighted the potential of the Unsupervised Environment Design framework, which generates environment instances/levels adaptively at the frontier of the agent’s capabilities using regret measures. While regret approaches have shown promise in generating feasible environments, they can produce difficult environments that are challenging for an RL agent to learn from. This is because regret represents the best-case (upper bound) learning potential and not the actual learning potential of an environment. To address this, we propose an alternative …

Evaluating Software Development Agents: Patch Patterns, Code Quality, And Issue Complexity In Real-World Github Scenarios, Zhi Chen, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

In recent years, AI-based software engineering has progressed from pre-trained models to advanced agentic workflows, with Software Development Agents representing the next major leap. These agents, capable of reasoning, planning, and interacting with external environments, offer promising solutions to complex software engineering tasks. However, while much research has evaluated code generated by large language models (LLMs), comprehensive studies on agent-generated patches, particularly in real-world settings, are lacking. This study addresses that gap by evaluating 4,892 patches from 10 top-ranked agents on 500 real-world GitHub issues from SWE-Bench Verified, focusing on their impact on code quality. Our analysis shows no single …

Backdoor Token Unlearning: Exposing And Defending Backdoors In Pretrained Language Models, Peihai Jiang, Xixiang Lyu, Yige Li, Jing Ma

Research Collection School Of Computing and Information Systems

Supervised fine-tuning has become the predominant method for adapting large pretrained models to downstream tasks. However, recent studies have revealed that these models are vulnerable to backdoor attacks, where even a small number of malicious samples can successfully embed backdoor triggers into the model. While most existing defense methods focus on post-training backdoor defense, efficiently defending against backdoor attacks during training phase remains largely unexplored. To address this gap, we propose a novel defense method called Backdoor Token Unlearning (BTU), which proactively detects and neutralizes trigger tokens during the training stage. Our work is based on two key findings: 1) …

Sticky Information Technology Investment: Theory And Empirical Evidence, Peng Liang, Hasan Cavusoglu, Nan Hu

Research Collection School Of Computing and Information Systems

This article provides a new way of thinking about managerial discretion in information technology (IT) investment decisions. We delve into the existence, antecedents, and consequences of sticky IT investment behavior, an understudied managerial deliberate resource commitment decision in response to changes in sales. Guided by downsizing theory, we initially theorize and find that IT investments exhibit stickiness: IT investments move downward less for sales decreases than they move upward for equivalent increases. Then drawing upon agency theory, adjustment costs theory, and managerial expectations theory—which influence managers’ motivation for downsizing—we predict and demonstrate that managers’ empire-building incentives, their avoidance of adjustment …

Beyond The Disaster Zone: How Disruptive Climate Event Shape Epa Oversight And Firm Behavior, Grace Fan

Research Collection School Of Accountancy

We examine how experiences with disruptive climate events impact the Environmental Protection Agency’s (EPA) monitoring activities and subsequent firm behaviors. Using plausibly exogenous variation in the exposure of EPA regional offices to major hurricanes, we find that exposed EPA regulators increase their monitoring efforts, measured by both inspection frequency and length, at facilities located in their jurisdictions but outside the disaster zone. This effect diminishes over time and is stronger for regulators located closer to the disaster area or experiencing the disaster for the first time. Importantly, the heightened regulatory monitoring following disasters has real effects on regulated facilities–they face …

Breaking Up Male-Dominated Collusive Arrangements: Women's Representation In Bureaucracy And Corruption, Beomgeun Cho, Heeun Kim

Research Collection School of Social Sciences

We investigate the impact of women's representation in bureaucracy on corruption. In this study, we demonstrate how women bureaucrats' experiences in male-dominated workplaces contribute to curbing corruption. To do this, we investigate three dimensions of women's representation (average share, average rank, and hierarchical dispersion) in South Korean regional governments. Empirical evidence shows that the higher women bureaucrats are ranked, the more bureaucrats are disciplined against misbehaviors and the lower the corruption risks are in public procurement. The findings imply that having women bureaucrats at higher ranks reduces corruption risks because they disrupt male-dominated collusive arrangements and abide by ethical standards …

Intellectual Humility Without Limits: Magnanimous Humility, Disagreement And The Epistemology Of Resistance, Brandon Yip

Research Collection School of Social Sciences

In this paper, I provide a characterisation of a neglected form of humility: magnanimous humility. Unlike most contemporary analyses of humility, magnanimous humility is not about limitations but instead presupposes that one possesses some entitlement in a context. I suggest that magnanimous intellectual humility (IH) consists in a disposition to appropriately refrain from exercising one's legitimate epistemic entitlements because one is appropriately motivated to pursue some epistemic good. I then shown that Magnanimous IH has an important role to play in contexts of disagreement and oppression. It calls on knowing parties to refrain from pressing their epistemic entitlements to facilitate …

Gender Biases Within Artificial Intelligence And Chatgpt: Evidence, Sources Of Biases, And Solutions, Qi Hui Jerlyn Ho, Andree Hartanto, Andrew Koh, Nadyanna M. Majeed

Research Collection School of Social Sciences

The growing adoption of Artificial Intelligence (AI) in various sectors has introduced significant benefits, but also raised concerns over biases, particularly in relation to gender. Despite AI's potential to enhance sectors like healthcare, education, and business, it often mirrors reality and its societal prejudices and can manifest itself through unequal treatment in hiring decisions, academic recommendations, or healthcare diagnostics, systematically disadvantaging women. This paper explores how AI systems and chatbots, notably ChatGPT, can perpetuate gender biases due to inherent flaws in training data, algorithms, and user feedback loops. This problem stems from several sources, including biased training datasets, algorithmic design …

Diversities Within Diversities, Sor-Hoon Tan

Research Collection School of Social Sciences

This chapter adds to the perspectives from Chinese philosophical traditions on the issues of ecological and economic inequalities by drawing attention to recent works on Daoism offering views on ecological problems that differ from Tongdong Bai’s chapter. It also highlights a common theme that is found in several chapters in this volume: the importance of relationship. Starting from Confucian relational ethics, it offers a different perspective from Bai and others who defend Confucian hierarchy by questioning the opposition between egalitarianism and hierarchy. It argues that Confucianism strongly objects to certain kinds of inequalities, which would put it in sympathy with …

Private Reachability Queries On Structured Encrypted Temporal Bipartite Graphs, Yulin Wu, Lanxiang Chen, Gaolin Chen, Yi Mu, Robert H. Deng

Research Collection School Of Computing and Information Systems

A temporal bipartite graph is a graph model that incorporates time-related information into its edges, making it suitable for modeling real-world phenomena like disease outbreaks. However, this temporal information is often sensitive. To protect the privacy of graph data, researchers have explored various approaches to preserve privacy in graph queries, with reachability queries being popular and fundamental as they determine the possibility of reaching one node from others in a graph. While privacy-preserving reachability queries have been extensively studied, existing efforts often overlook the valuable attribute information present in both edges and nodes of the graphs. Moreover, reachability queries on …

Cachealarm: Monitoring Sensitive Behaviors Of Android Apps Using Cache Side Channel, Jianwen Tian, Haoyu Ma, Debin Gao, Xiaohui Kuang

Research Collection School Of Computing and Information Systems

Malware attack has been a serious threat to the security and privacy of both individual and corporation users of the Android platform. Business entities seek to protect themselves by means of monitoring privacy-related sensitive behaviors conducted on company-issued Android devices. However, due to Android’s own access control and privacy protection policies, this is difficult to be done with third-party apps using only normal privileges. Existing works proposed using side-channel readings from leaky APIs and system virtual files to speculate runtime app behaviors, which could be unreliable due to future system updates (that ban exploited resources), hardware jittering, etc. In this …

Personamagic: Stage-Regulated High-Fidelity Face Customization With Tandem Equilibrium, Xinzhe Li, Jiahui Zhan, Shengfeng He, Yangyang Xu, Junyu Dong, Huaidong Zhang, Yong Du

Research Collection School Of Computing and Information Systems

Personalized image generation has made significant strides in adapting content to novel concepts. However, a persistent challenge remains: balancing the accurate reconstruction of unseen concepts with the need for editability according to the prompt, especially when dealing with the complex nuances of facial features. In this study, we delve into the temporal dynamics of the text-to-image conditioning process, emphasizing the crucial role of stage partitioning in introducing new concepts. We present PersonaMagic, a stage-regulated generative technique designed for high-fidelity face customization. Using a simple MLP network, our method learns a series of embeddings within a specific timestep interval to capture …

An Aspect Performance-Aware Hypergraph Neural Network For Review-Based Recommendation, Junrui Liu, Tong Li, Di Wu, Zifang Tang, Yuan Fang, Zhen Yang

Research Collection School Of Computing and Information Systems

Online reviews allow consumers to provide detailed feedback on various aspects of items. Existing methods utilize these aspects to model users' fine-grained preferences for specific item features through graph neural networks. We argue that the performance of items on different aspects is important for making precise recommendations, which has not been taken into account by existing approaches, due to lack of data. In this paper, we propose an aspect performance-aware hypergraph neural network (APH) for the review-based recommendation, which learns the performance of items from the conflicting sentiment polarity of user reviews. Specifically, APH comprehensively models the relationships among users, …

On Strategy-Proofness And The Salience Of Single-Peakedness In A Private Goods Allotment Problem, Shurojit Chatterji, Jordi Masso, Shigehiro Serizawa

Research Collection School Of Economics

We consider strategy-proof rules operating on a rich domain of preference profiles in a set up where multiple private goods have to be assigned to a set of agents with entitlements and where preferences display satiation. We show that if the rule is in addition tops-only, same-sided and individually rational with respect to the entitlements, then the preferences in the domain have to satisfy a variant of single-peakedness (referred to as multi-dimensional single-peakedness relative to the entitlements). We also provide a converse of this main finding. It turns out that this domain coincides with the one already identified in a …

Food Self-Sufficiency And Building-Integrated Urban Agriculture: Lessons From Singapore, Tomoki Fujii, Christoph Waibel, Xinyi Du, Zhongming Shi

Research Collection School Of Economics

Singapore aims to achieve 30 percent food self-sufficiency by 2030, known as the 30-by-30 target, and this paper reviews Singapore’s changing landscape for urban agriculture and its recent progress towards the 30-by-30 goal, highlighting key challenges such as land constraints, high production costs, and resource limitations. Building-integrated agriculture (BIA), which utilizes building surfaces such as rooftops, façades, and balconies for food production within urban environments, is examined as a potential way to increase self-sufficiency through a simulation. Despite the BIA’s potential, practical issues—including regulatory concerns and infrastructure limitations—must be addressed before it can be implemented at scale. Insights from Singapore’s …

Fintech’S Rise Reshaping Asean’S Financial Future, Swee Liang Tan

Research Collection School Of Economics

Fintech can offset the diminishing effects of bank lending on economic growth. By leveraging innovative technologies, fintech companies can reduce transaction costs, address information asymmetry issues and streamline processes such as customer evaluation and risk assessment. This strategic collaboration between banks and fintech firms can lead to improved efficiency and effectiveness in financial services, such as pressuring traditional banks to innovate in areas such as customer interface and processes, ultimately boosting economic growth.

Unpacking Singapore’S Leasehold Relativity Table: An Empirical And Legal Analysis, Koon Shing Kwong, Jing Rong Goh, Seng Wei, Edward Ti

Research Collection School Of Economics

In Singapore, most land is state-owned, with the state generally issuing leasehold estates via state leases of not more than 99 years1, depending on the intended land use. Naturally, the value of a leasehold estate, which erodes over time as the lease approaches the end of its term, is a key component of the premium charged for lease renewals, or the tax imposed for permission given in relation to a development that would increase the value of the land. By law, the state valuation of leasehold land is prescribed by a leasehold relativity table colloquially known as ‘Bala’s Curve’ or …

Deepvmunprotect: Neural Network-Based Recovery Of Vm‑Protected Android Apps For Semantics‑Aware Malware Detection, Xin Zhao, Mu Zhang, Xiaopeng Ke, Yu Pan, Yue Duan, Sheng Zhong, Fengyuan Xu

Research Collection School Of Computing and Information Systems

The emerging virtual machine-based Android packers render existing unpacking techniques ineffective. The state-of-the-art unpacker falls short because it relies on unreliable heuristics and manually crafted semantic models. Hence, it cannot precisely recover app semantics necessary for malware detection. In this paper, we propose DeepVMUnProtect, a deep learning-based approach to automatically and accurately capture the semantics of VM-packed code, so as to facilitate semantic-based Android malware classification. Experiments have shown that DeepVMUnProtect outperforms the state-of-the-art tool on recovering opcode semantics in Qihoo(58.3%), Baidu(47.5%) and NMMP (58.8%) respectively, and can enable semantics-aware malware detection which prior work fails to do.