Digital Commons Network^™

Scaling Up Multi-Agent Reinforcement Learning For Large Agent Teams And Long-Horizon Tasks: A Survey, Minghong Geng, Shubham Pateria, Budhitama Subagdja, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Multi-agent reinforcement learning (MARL) empowers multiple autonomous agents to acquire effective policies for collaborative problem-solving. Over the last decade, MARL has seen significant advancements, with numerous algorithms achieving impressive performance across various benchmarks and real-world applications. Nevertheless, the scalability of multi-agent systems, in terms of the number of agents and the length of the task horizon, remains a critical consideration for applying MARL methods to complex problem-solving. Given that a dedicated review of the existing approaches and challenges in scaling up multi-agent systems remains largely absent, this survey aims to bridge this gap by delivering a comprehensive review of MARL …

Rode: Linear Rectified Mixture Of Diverse Experts For Food Large Multi-Modal Models, Pengkun Jiao, Xinlan Wu, Bin Zhu, Jingjing Chen, Chong-Wah Ngo, Yu-Gang

Research Collection School Of Computing and Information Systems

Large Multi-modal Models (LMMs) have significantly advanced a variety of vision-language tasks. The scalability and availability of high-quality training data play a pivotal role in the success of LMMs. In the realm of food, while comprehensive food datasets such as Recipe1M offer an abundance of ingredient and recipe information, they often fall short of providing ample data for nutritional analysis. The Recipe1M+ dataset, despite offering a subset for nutritional evaluation, is limited in the scale and accuracy of nutrition information. To bridge this gap, we introduce Uni-Food, a unified food dataset that comprises over 100,000 images with various food labels, …

Benchmarking Gaslighting Negation Attacks Against Multimodal Large Language Models, Bin Zhu, Yinxuan Gui, Huiyan Qi, Jingjing Chen, Chong-Wah Ngo, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Multimodal Large Language Models (MLLMs) have exhibited remarkable advancements in integrating different modalities, excelling in complex understanding and generation tasks. Despite their success, MLLMs remain vulnerable to conversational adversarial inputs. In this paper, we systematically study gaslighting negation attacks—a phenomenon where models, despite initially providing correct answers, are persuaded by user-provided negations to reverse their outputs, often fabricating justifications. We conduct extensive evaluations of state-of-the-art MLLMs across diverse benchmarks and observe substantial performance drops when negation is introduced. Notably, we introduce the first benchmark GaslightingBench, specifically designed to evaluate the vulnerability of MLLMs to negation arguments. GaslightingBench consists of multiple-choice …

Not Too Early, Not All At Once: Design Tensions In Ai-Mediated Self-Disclosure In Online Dating, Pei-Hua Tsai, Tianyi Zhang, Emran Bin Elias Poh, Anthony Tang, Yung-Ju Chang

Research Collection School Of Computing and Information Systems

Online dating relies on self-disclosure, yet initial conversations are fragile: users must navigate uncertainty around timing, boundaries, and reciprocity with little shared context. While advances in AI raise the possibility of mediating disclosure, how such support might reshape the experience of early-stage relational disclosure remains underexplored. We conducted 29 semi-structured interviews to examine how daters envision AI-mediated self-disclosure in online dating. Our findings surface recurring design tensions rather than simple opportunities or risks. Participants welcomed guidance that could pace disclosure, support reflection, and reduce social awkwardness, but stressed preserving agency and authorship. They valued interpretive assistance for sense-making of ambiguous …

Language Embeddings Meet Shallow Autoencoders, Rodrigo Alves, Vojtěch Vančura, Pavel Kordík, Antoine Ledent

Research Collection School Of Computing and Information Systems

Shallow autoencoders are appealing recommenders due to their simplicity, scalability, and competitive retrieval quality, but they struggle in strict cold-start settings where new items have no interactions. We propose an inductive shallow autoencoder that leverages item side information (language embeddings) by fixing the decoder to item features and learning only an encoder in the same semantic space. To prevent trivial self-reconstruction without enforcing a hard zero diagonal, we introduce diagonal gating: a leave-one-item-out objective that blocks the self-copy shortcut only for the item being updated while retaining context from the rest of the user history. An alternating-style optimization trains the …

Sok: Understanding Zkvm: From Research To Practice, Guomin Yang, Yunbo Yang, Yuejia Cheng, Haibo Tang, Bingsheng Zhang, Kui Ren

Research Collection School Of Computing and Information Systems

Zero-knowledge virtual machine (zkVM) is a powerful infrastructure for proving the correctness of a program execution with a succinct proof, attracting significant interest from researchers, developers, and users. It has been widely used in applications such as blockchain rollups, privacy-preserving machine learning, and off-chain computation. As the field grows, a wide range of zkVMs have been proposed. However, they adopt different choices in instruction formats, trace layouts, and proving backends, which results in a highly heterogeneous design landscape and makes it difficult to understand the relations among these systems.To bridge this gap, we provide a comprehensive study of zkVMs that …

Anatomical Domain Shifts: Test-Time Heterogeneous Adaptation For 3d Human Pose Prediction, Qiongjie Cui, Pan Zhou, Jingjing Chen, Na Zhao

Research Collection School Of Computing and Information Systems

The research frontier in human pose prediction (HPP) is advancing toward continual test-time adaptation (TTA), where models must self-adapt to dynamic test distributions. To date, the homeostatic continual TTA remains the sole viable solution, which isolates the model parameters and update domain-sensitive ones. Despite mitigating full-body domain gaps, human anatomical heterogeneity (domain shifts often localize to specific regions) is ignored. This anatomical-agnostic approach forces uniform parameter adaptation across kinematically distinct segments, causing: over-adaptation of stable regions and under-adaptation of shift-prone articulations. To address it, we introduce TT-HA, a novel Test-Time Heterogeneous Adaptation that implicitly estimates domain changes for anatomical segments, …

Enhancing Pointing Gestures Of Non-Hmd Users In Asymmetric Collocated Mixed Reality Collaboration, Nam-Dang Vo, Van-Vinh Thai, Anthony Tang, Khanh-Duy Le

Research Collection School Of Computing and Information Systems

A common collocated group setting in mixed-reality (MR) collaboration is a person wearing a MR headset (HMD user) and presenting MR contents to audiences who are not provided with such specialized devices (Non-HMD users). In this setting, while Non-HMD users can view the MR environment shown on a large physical display, it still remains challenging for the HMD user to interpret their pointing gesture when they spatially refer to objects in the MR environment. To address this, we designed and evaluated two pointing techniques—SCREEN and SCREEN+SPACE—that support Non-HMD users in referring to MR content. Screen pointing allows users to refer …

Happycal: Designing Text And Image-Based Supports For Savouring Positive Work Experiences, Molly Stewart, Minghao Cai, Anthony Tang, Sam Liu, Chris Mosunic, Sowmya Somanath

Research Collection School Of Computing and Information Systems

Savouring positive work experiences can promote positive affect and well-being at work, yet there is limited guidance on how digital applications can support workers to engage in savouring. We developed HappyCal, a work-focused savouring application offering two forms of savouring support: text-based, a common modality in workplace reflection tools, and images, a largely unexplored approach in work-related savouring. We conducted an exploratory qualitative study where participants (N=36) used HappyCal over five days and engaged in savouring through either a text-only modality (n=17) or text input paired with image output (n=19). We found that (1) participants in both groups reported heightened …

How Do Machine Learning Models Change?, Joel Castaño, Rafael Cabañas, Antonio Salmerón, David Lo, Silverio Martínez-Fernández

Research Collection School Of Computing and Information Systems

The proliferation of Machine Learning (ML) models and their open source implementations has transformed AI research and applications. Platforms like Hugging Face (HF) enable this evolving ecosystem, yet a large-scale longitudinal study of how these models change is lacking. This study addresses this gap by analyzing over 680,000 commits from 100,000 models and 2,251 releases from 202 of these models on HF using repository mining and longitudinal methods. We apply an extended ML change taxonomy to classify commits and use Bayesian networks to model temporal patterns in commit and release activities. Our findings show that commit activities align with established …

Ai In Healthcare: Regulatory Guidelines And Judge-Made Negligence Principles For Ai Implementers, Gary K. Y. Chan

Research Collection Yong Pung How School Of Law

The use of artificial intelligence (AI) in healthcare may, notwithstanding its potential benefits, result in harm to patients from allegedly negligent acts or omissions by hospitals and medical doctors. In such circumstances, how should the principles in the tort of negligence (duty of care, breach, causation, remoteness of damage, and defences) respond to AI innovations in healthcare? In particular, how may the standard of care expected of hospitals and medical doctors be informed by regulatory guidelines? We refer to case law precedents and regulatory guidelines on the roles and responsibilities of doctors and hospitals as AI implementers. Importantly, they prompt …

International Advice, Mature Democracies And The Venice Commission, Maartje De Visser

Research Collection Yong Pung How School Of Law

The aim of this article is to explore the role and practice of constitutional advice-giving as it relates to mature democracies. More precisely, this article considers how the Venice Commission – the official advisory body for constitutional matters of the Council of Europe – exercises its mandate vis-à-vis such democracies through the delivery of country-specific Opinions, based on a close reading of all such Opinions rendered between 2002 and 2024. It examines the substantive focus of those Opinions, the way they engage with constitutional experiences elsewhere, as well as the tone of the assessment and the approach taken in formulating …

Social Exclusion Cues And Collective Action Motivation In Environmental Campaigns: The Buffering Role Of Social Support, Hwan-Ho Noh, Warren B. Chik

Research Collection Yong Pung How School Of Law

This study examines the influence of social exclusion cues on public engagement with environmental campaigns on social media, focusing on how such cues affect advertising attitudes and collective action motivation. Across two experimental studies, we investigated whether low engagement metrics, such as a small number of likes, function as symbolic exclusion cues that reduce perceived public support. The results indicate that social exclusion cues negatively affect advertising attitudes and collective action motivation, ultimately decreasing the intention to engage with the campaign. However, the presence of social support cues, which signal sustained public interest and social endorsement, effectively mitigated these negative …

Not Retirement, But A Rewiring And Fresh Perspectives Post-Dbs, Says Piyush Gupta [Interview With Business Times], Su Shyan Lee, Piyush Gupta

Oral History Collection

In an interview with The Business Times, SMU Chairman Piyush Gupta discussed topics ranging from meditation and conservation to the future of education. On his role at SMU, he emphasised his desire to help shape tertiary education and noted that the rise of AI makes it critical to rethink how we learn. Mr Gupta believes that in an AI-driven world, students must move beyond domain knowledge to focus on interpreting information, "connecting the dots," and engaging others with genuine empathy. He is supportive of a more experiential curriculum at SMU — one that focuses on regional exposure through internships and …

Sequential Robustness In Adversarial Reinforcement Learning, Roman Lok-Ming Belaire

Dissertations and Theses Collection (Open Access)

My goal is to build autonomous systems that expand the reach of human capability in challenging domains such as undersea and space exploration, disaster response, and large-scale infrastructure. In everyday settings, these systems will increasingly appear in safety-critical applications such as autonomous driving, robotics, and industrial manufacturing. A central requirement for these systems is the ability to operate reliably under uncertainty, particularly when the environment behaves in unanticipated ways.

The robust handling of unforeseen environment dynamics is therefore a technical cornerstone of autonomous decision-making; Adversarial attacks provide a useful and principled lens through which to study this problem. Adversarial \textit{robustness}, …

Factors Influencing Patient Recall For Follow-Up In The Multi-Stage Deployment Of Medical Intelligent Screening Systems, Wenjie Chen

Dissertations and Theses Collection (Open Access)

With the deep integration of artificial intelligence (AI) technology in the healthcare sector, intelligent screening systems have achieved significant breakthroughs in predictive accuracy; however, they commonly face a management challenge in clinical implementation: accurate early warnings but limited patient recall. To address this ineffective technological empowerment issue in practice, this study proposes a theoretical framework featuringbidirectional interactions between "technological empowerment" and"relational empowerment" based on empowerment theory to systematicallyinvestigate the dynamic evolution mechanisms and boundary conditions of patient recall willingness under multi-stage, cross-departmental deployment of medical intelligent screening systems. Using a leading municipal hospital as the primary experimental setting, the study …

Perspectives On Interpretability For Neural Text Representations, Jia Peng Lim

Dissertations and Theses Collection (Open Access)

In this dissertation, we investigate interpretability in the three elements of learning neural text representations: inputs, passed into models, to produce probabilistic outputs. We emphasise perspectives as we present alternative novel methods to mine and organise meaning in this work.

Models. We initiate our investigation by examining Neural Topic Models (NTM), proposing an alternate angle of interpreting its word-topic distribution, producing better topic representations for interpretation. Our method maps the problem of finding these better interpretations to classical NP-hard graph problems, enabling examination of topic distributions in a composite manner. Next, we apply our previous findings to extract interpretations from …

The Impact Of Successor Differences In Family Businesses On Firm Performance: An Empirical Analysis Based On China Listed Company Data, Xulong Zhao

Dissertations and Theses Collection (Open Access)

Family businesses are an important pillar of global economic development. Currently, Chinese private enterprises are ushering in an unprecedented wave of intergenerational succession. At this critical historical juncture, "who" takes over and how the succession behavior affects firm performance have become core issues of common concern to academia and practice. Traditional principal-agent theory usually posits that introducing professional managers can break the limitations of family governance and improve corporate efficiency. However, Chinese family businesses are deeply rooted in specific institutional environments and cultural soils, where blood-based "relational trust" and socioemotional wealth (SEW) play irreplaceable roles in power transitions. Therefore, how …

Reflections On Building Data Harnessing Literacy: A Three-Tier Approach For Business, Redzuan Abdullah

Research Collection Library

This is part of a summary of my presentation at the INCONECSS 2025, on SMU Libraries’ approach to offering data discovery services to its research community. As academic libraries evolve into vital research partners, business librarians face unprecedented demands to support complex, data-intensive research across financial and business databases. This piece explores a fundamental question: How do we keep up?

Func: Reducing The Impact Of Android Framework Evolution On Malware Detection, Hailong Yu, Tiantian Wang, Lwin Khin Shar, Hanmeng Li, David Lo

Research Collection School Of Computing and Information Systems

Android malware detection approaches commonly use APIs and permissions as features for classifying malware. However, since the release of the first Android operating system in 2008, the Android framework has undergone numerous version updates. The evolution of the Android framework over time has led to changes in APIs and permissions, including deprecations and replacements. These changes can result in inaccurate characterization of Android malware, thereby affecting performance of malware detectors. There is a lack of methods to mitigate the impact of Android framework evolution on malware detection. To fill this gap, we conduct a systematic study of the impact of …

Benchmarking Gaslighting Attacks Against Speech Large Language Models, Jinyang Wu, Bin Zhu, Xiandong Zou, Qiquan Zhang

PhD Student’s Publications Collection

As Speech Large Language Models (Speech LLMs) become increasingly integrated into voice-based applications, ensuring their robustness against manipulative or adversarial input becomes critical. Although prior work has studied adversarial attacks in text-based LLMs and vision-language models, the unique cognitive and perceptual challenges of speech-based interaction remain underexplored. In contrast, speech presents inherent ambiguity, continuity, and perceptual diversity, which make adversarial attacks more difficult to detect. In this paper, we introduce gaslighting attacks, strategically crafted prompts designed to mislead, override, or distort model reasoning as a means to evaluate the vulnerability of Speech LLMs. Specifically, we construct five manipulation strategies: Anger, …

Teacher-Student Diffusion Model For Text-Driven 3d Hand Motion Generation, Ching Lam Cheng, Bin Zhu, Shengfeng He

PhD Student’s Publications Collection

Generating realistic 3D hand motion from natural language is vital for VR, robotics, and human-computer interaction. Existing methods either focus on full-body motion, overlooking detailed hand gestures, or require explicit 3D object meshes, limiting generality. We propose TSHaMo, a model-agnostic teacher-student diffusion framework for text-driven hand motion generation. The student model learns to synthesize motions from text alone, while the teacher leverages auxiliary signals (e.g., MANO parameters) to provide structured guidance during training. A co-training strategy enables the student to benefit from the teacher’s intermediate predictions while remaining text-only at inference. Evaluated using two diffusion backbones on GRAB and H2O, …

Digital Grief Technology To Support Bereavement: A Systematic Review Of Potential Benefits And Risks, Xun Ci Soh, Adalia Yin Hui Goh, Paye Shin Koh, Andree Hartanto

Research Collection School of Social Sciences

Grief is a universal and inevitable experience. However, the way we support the bereaved is changing, especially in the digital era. This systematic review examines the potential benefits and risks associated with various digital grief technologies, including online grief support groups, generative AI chatbots, online memorials, online therapy interventions, virtual reality, and digitally reproduced visuals or audio of the deceased. A systematic search was conducted in seven databases, and 30 articles were included in the final review. Findings indicate that digital grief technologies offer several benefits, such as reductions in grief and depressive symptoms, enhanced social support, greater accessibility, and …

Understanding Critical Thinking In Generative Artificial Intelligence Use: Development, Validation, And Correlates Of The Critical Thinking In Ai Use Scale, Gabriel R. Lau, Wei Yan Low, Louis Tay, Ysabel Thereze Ang Guevarra, Dragon Gašević, Andree Hartanto

Research Collection School of Social Sciences

Generative AI tools are increasingly embedded in everyday work and learning, yet their fluency, opacity, and propensity to hallucinate mean that users must critically evaluate AI outputs rather than accept them at face value. The present research conceptualises critical thinking in AI use as a dispositional tendency to verify the source and content of AI-generated information, to understand how models work and where they fail, and to reflect on the broader implications of relying on AI. Across six studies ( N = 1341), we developed and validated the 13-item critical thinking in AI use scale and mapped its nomological network. …

Insolvency Law In The Global South: Lessons For The Global North, Aurelio Gurrea-Martinez

Research Collection Yong Pung How School Of Law

Despite the influence of the Global North in many insolvency laws and practices in the Global South, this article shows that the Global South has innovated in many aspects of insolvency law. In some cases, these innovations consist of solutions that, with certain adjustments, have been imported from the Global North. In others, they are really ‘autochthonous innovations’ from the Global South. This article identifies both types of innovations, providing examples from jurisdictions such as Brazil, Chile, China, Colombia, Dominican Republic, India, Malaysia, Mexico, Myanmar, Peru, Philippines, Thailand and Uruguay. More importantly, it will be shown how those innovations from …

Bitcoin Options Risk-Reversal Predictability, Meng Hwee Neo

Dissertations and Theses Collection (Open Access)

This dissertation examines whether Bitcoin options risk-reversal (RR) spreads predict future Bitcoin returns, using daily Deribit data from April 2021 to December 2025 (1,723 observations).

In the primary 25-delta, 90-day specification, the RR coefficient is significant at the 1% level in a thirteen-variable baseline regression with Newey–West standard errors. Robustness tests across all available delta–tenor specifications show statistically significant RR coefficients in the majority of configurations, concentrating in the 30–180-day maturity band.

The butterfly spread (BF) also predicts returns. Both remain significant after controls are added: the RR t-statistic increases from 1.91 (univariate) to 3.73 (full baseline), and the BF …

Research On The Spatial Diffusion Process And Influencing Factors Of Traditional Chinese Medicine Brands: Evidence From Beijing Tong Ren Tang, Yuehua Shen

Dissertations and Theses Collection (Open Access)

Against the backdrop of the dual drivers of the revitalization and development of traditional Chinese medicine (TCM) and the rising demand for health-related consumption, the nationwide expansion of TCM chain brands has become a prominent issue of shared concern for both industry and policy circles. However, existing research largely treats cities as mutually independent units and overlooks the spatial externalities arising from geographic proximity. It also implicitly assumes that the driving factors remain stable over a brand's expansion life cycle. This makes it difficult to capture the dynamic evolution of the logic governing store expansion across different stages. Accordingly, this …

Essays On High-Frequency Dynamics Of Asset Prices, Yuhong Zhu

Dissertations and Theses Collection (Open Access)

This dissertation studies econometric inference for high-frequency financial data, with a focus on detecting nonstandard drift and volatility dynamics in continuous-time models.

The first chapter proposes a new framework for uniform inference on explosive drift in high-frequency data, where conventional Gaussian approximations can fail due to the non-Gaussian behavior of short-window spot statistics. Under fixed-window asymptotics, these statistics are coupled with dependent t variables, and their maximum converges to a Fréchet distribution. We establish an anti-clustering condition for dependent t-statistics under overlapping windows and develop a feasible coupling-based test. Simulation results demonstrate better size control, and the empirical findings suggest …

Technological Strategy Paths, Dynamic Capability And Innovation Performance In Complex Product Industries: A Mechanism Study Based On China’S New Energy Vehicle Industry, Fan Zhang

Dissertations and Theses Collection (Open Access)

In recent years, the new energy vehicle (NEV) industry has developed rapidly, moving from electrification toward an intelligent phase characterized by software-defined vehicles. NEV firms pursue product and technological innovation through diverse strategic paths, including internal research and development, collaborative development, and platform-based development, yet their innovation performance varies. Most extant research focuses on the industry level, offering limited insight into how, at the firm level, different technological paths shape the innovation outcomes through firms’ dynamic capabilities. It still remains unclear whether and how these relationships among paths, capabilities, and innovation outcomes differ across technological domains.

This study adopts a …

The Seller Valuation Of Family Firm Exits: Socioemotional Wealth And Ceo Equity Ownership, Masahiro Nishii

Dissertations and Theses Collection (Open Access)

This study examines seller valuation in family-firm exit using the initial asking price from private M&A sales. In the sale of a family firm, it is known that seller valuation may go beyond financial value alone because owners often hold strong emotional attachments to the firm. However, it remains unclear whose judgment brings that non financial value into seller valuation and why this varies across types of family CEOs.

Drawing on socioemotional wealth theory, the study develops the mechanism at the CEO level. I argue that the anticipated SEW loss translates into seller valuation through the CEO’s judgment. When the …