Digital Commons Network^™

Benchmarking Gaslighting Negation Attacks Against Reasoning Models, Bin Zhu, Hailong Yin, Jingjing Chen, Yu Gang Jiang

Research Collection School Of Computing and Information Systems

Recent advances in reasoning-centric models promise improved robustness through mechanisms such as chain-of-thought prompting and test-time scaling. However, their ability to withstand gaslighting negation attacks—adversarial prompts that confidently deny correct answers—remains underexplored. In this paper, we conduct a systematic evaluation of three state-of-the-art reasoning models, i.e., OpenAI’s o4-mini, Claude-3.7-Sonnet and Gemini-2.5-Flash, across three multimodal benchmarks: MMMU, MathVista, and CharXiv. Our evaluation reveals significant accuracy drops (25–29% on average) following gaslighting negation attacks, indicating that even top-tier reasoning models struggle to preserve correct answers under manipulative user feedback. Built upon the insights of the evaluation and to further probe this vulnerability, …

Realign: Text-To-Motion Generation Via Step-Aware Reward-Guided Alignment, Wanjiang Weng, Xiaofeng Tan, Junbo Wang, Guo-Sen Xie, Pan Zhou, Hongsong Wang

Research Collection School Of Computing and Information Systems

Text-to-motion generation, which synthesizes 3D human motions from text inputs, holds immense potential for applications in gaming, film, and robotics. Recently, diffusion-based methods have been shown to generate more diversity and realistic motion. However, there exists a misalignment between text and motion distributions in diffusion models, which leads to semantically inconsistent or low-quality motions. To address this limitation, we propose Reward-guided sampling Alignment (ReAlign), comprising a step-aware reward model to assess alignment quality during the denoising sampling and a reward-guided strategy that directs the diffusion process toward an optimally aligned distribution. This reward model integrates step-aware tokens and combines a …

Airaclex: Automated Detection Of Price Oracle Manipulations Via Llm-Driven Knowledge Mining And Prompt Generation, Bo Gao, Yuan Wang, Qingsong Wei, Yong Liu, Rick Siow Mong Goh, David Lo

Research Collection School Of Computing and Information Systems

Decentralized finance (DeFi) applications depend on accurate price oracles to ensure secure and fair transactions. However, poorly integrated oracles remain susceptible to manipulation, enabling attackers to exploit smart contract logic for unfair asset valuation and financial gain. While many such vulnerabilities are only detected after deployment, smart contracts are typically immutable once deployed, making post-hoc fixes costly or infeasible. This highlights the critical need for detecting oracle manipulation risks before deployment. In this paper, we propose AiRacleX, a novel LLM-driven framework that enables pre-deployment detection of price oracle manipulation vulnerabilities by leveraging the complementary strengths of multiple large language models …

Freedom Of Expression Protection And Corporate Concealment Of Bad News: Evidence From State Anti-Slapp Laws, Jimmy Lee, Shaphan Ng, Il Sun Yoo, Liandong Zhang

Research Collection School Of Accountancy

The protection of free speech enhances the ability of various public stakeholders to disseminate privately observed adverse information about public firms, making it difficult for corporate managers to conceal negative information about their companies. Using the staggered enactment of anti strategic lawsuit against public participation (anti-SLAPP) laws across U.S. states as a shock that strengthens free speech protection, we show that stronger protection is associated with less concealment of bad news. This is evidenced by a lower likelihood of stock price crashes, a decreased probability of accounting fraud, and an increased frequency of firm-initiated negative press releases. These results are …

‘Salami Slicing’ And Issue Estoppel: Foreign Decisions On The Governing Law, Adeline Chong

Research Collection Yong Pung How School Of Law

Whether an issue estoppel arises over foreign decisions on the governing law of the claim has not been directly considered by an English court, but decisions in other jurisdictions show that this question is increasingly being raised in litigation. Is there identity of issue if the two courts apply different choice of law rules? The answer turns on whether a broad or narrow framing of the issue is adopted. It is suggested that, absent an issue which is subject to forum international public policy, forum overriding mandatory rules or which is one that the forum court retains the prerogative to …

Security-Enhanced Decentralized Conditional Privacy-Preserving Authentication In Vanets, Suqin Luo, Xinghua Li, Yinbin Miao, Xuelin Cao, Zhan Zhang, Yunwei Wang, Deng R.H.

Research Collection School Of Computing and Information Systems

To ensure the legitimacy of communicators while ad dressing the privacy concerns of vehicles in vehicular ad-hoc networks (VANETs), conditional privacy-preserving authentication (CPPA) schemes have been proposed. Given that existing schemes suffer from single point of failure due to centralized authorities, several distributed CPPA schemes have been proposed. However, these schemes all ignore the tight cementation between system secret keys and the authority, which could be a serious threat to system security, that the compromised authority may leak the system secret key. To address these issues, we propose a security enhanced decentralized conditional privacy-preserving authentication (DCPPA) scheme. DCPPA first introduces …

Tempo: Training-Time Equilibration Of Modalities For Per-Sample Optimization In Multimodal Sentiment, Yi Zhao, Erik Cambria, Xiaosong E, Xianxun Zhu

Research Collection School Of Computing and Information Systems

Multimodal sentiment models often become over-reliant on the “easiest” modality (typically text), leading to three coupled sub-problems: (i) representation-level dominance, where weaker modalities contribute little to the fused representation; (ii) optimization-level dominance, where the strongest modality drives most gradient updates and suppresses learning in others; and (iii) robustness degradation, where audio or vision fail under noise or missing inputs at test time. We present TEMPO, a plug-and-play training framework that mitigates these issues by rebalancing learning pressure across modalities while leaving inference unchanged. For each mini-batch, TEMPO estimates relative modality strength and applies two synchronized, training-only controls: selective forward attenuation …

Nondeterministic Polynomial-Time Problem Challenge: An Ever-Scaling Reasoning Benchmark For Llms, Chang Yang, Ruiyu Wang, Junzhe Jiang, Qi Jiang, Qinggang Zhang, Yanchen Deng, Shuxin Li, Shuyue Hu, Bo Li, Florian T. Pokorny, Xiao Huang, Xinrun Wang

Research Collection School Of Computing and Information Systems

Reasoning is the fundamental capability of large language models (LLMs). Due to the rapid progress of LLMs, there are two main issues of current benchmarks: i) these benchmarks can be crushed in a short time (less than 1 year), and ii) these benchmarks may be easily hacked. To handle these issues, we propose the ever-scalingness for building the benchmarks which are scaling over complexity against crushing, instance against hacking and exploitation, oversight for easy verification, and coverage for real-world relevance. This paper presents Nondeterministic Polynomial-time Problem Challenge (NPPC), an ever-scaling reasoning benchmark for LLMs. Specifically, the NPPC has three main …

Reliable-Data-Split (Rds): Maximizing Model Potential With Reinforced Selection Strategy, Hoang D. Nguyen, Xuan-Son Vu, Quoc Tuan Truong, Duc-Trong Le

Research Collection School Of Computing and Information Systems

The nexus between data characteristics and parametric models is fundamental for developing effective and reliable artificial intelligence (AI) systems. Mismatches in data properties for model development may lead to deleterious effects on AI model performance in machine learning practice. This paper proposes a Reliable Data Split (RDS) procedure to learn how to select data points that will generalise the target domain adequately by employing prior knowledge of the data generative process. We introduce a reinforced selection strategy using deep reinforcement learning with diverse black box predictors in maximising ensemble rewards as the proxy of model performance potential while maintaining an …

A Learning‑Augmented Dynamic Programming Approach For Orienteering Problem With Time Windows, Guansheng Peng, Lining Xing, Fuyan Song Ma, Aldy Gunawan, Aldy Gunawan

Research Collection School Of Computing and Information Systems

Recent years have witnessed a surge of interest in solving combinatorial optimization problems (COPs) using machine learning techniques. Motivated by this trend, we propose a learning-augmented exact approach for tackling an NP-hard COP, the Orienteering Problem with Time Windows, which aims to maximize the total score collected by visiting a subset of vertices in a graph within their time windows. Traditional exact algorithms rely heavily on domain expertise and meticulous design, making it hard to achieve further improvements. By leveraging deep learning models to learn effective relaxations of problem restrictions from data, our approach enables significant performance gains in an …

Semi‑Supervised Graph Anomaly Detection Via Robust Homophily Learning, Guoguo Ai, Hezhe Qiao, Hui Yan, Guansong Pang

Research Collection School Of Computing and Information Systems

Current semi-supervised graph anomaly detection (GAD) methods utilizes a small set of labeled normal nodes to identify abnormal nodes from a large set of unlabeled nodes in a graph. These methods posit that 1) normal nodes share a similar level of homophily and 2) the labeled normal nodes can well represent the homophily patterns in the entire normal class. However, this assumption often does not hold well since normal nodes in a graph can exhibit diverse homophily in real-world GAD datasets. In this paper, we propose RHO, namely Robust Homophily Learning, to adaptively learn such homophily patterns. RHO consists of …

Imprisonment When An Offender Cannot Pay A Fine, Benjamin Joshua Ong

Research Collection Yong Pung How School Of Law

According to a common-law rule in place since the 1993 case of Low Meng Chay v Public Prosecutor [1993] 1 SLR(R) 46, if the court is minded to impose a fine but the offender will clearly be unable to pay a fine, the offender should be sentenced to imprisonment instead (as opposed to a fine coupled with a default imprisonment term). While one can understand why the courts may apply this practice, the practice obscures the crucial distinction between: (a) being sentenced to a fine, then imprisoned in default of payment (which, it is submitted, is the correct course of …

A Study On The Influence Of Entity Internet Hospital Platforms And Health Literacy Facilitation On Postoperative Patients’ Platform Usage, Hong Ye

Dissertations and Theses Collection (Open Access)

Entity Internet Hospital (EIH) is emerging online medical service platforms in China. Currently, while EIHs in China are developing rapidly, their usage rates remain low. However, postoperative patients have substantial follow-up consultation needs. Consequently, incentivizing postoperative patients to use EIH from the demand side is a highly challenging proposition in the current era. Grounded in Technology Acceptance Model and Health Belief Model, this study examined the impacts of platform and health literacy facilitation on platform usage (including intention and frequency) among 631 postoperative patients with cardiovascular disease from four medical institutions in Wuhan, and explored the moderating role of literacy …

Corporate Strategic Changes In Response To Negative Performance Gaps: The Role Of Listing Board In China, Lei Yu

Dissertations and Theses Collection (Open Access)

This dissertation investigates how Chinese listed companies respond strategically to negative performance gaps under different institutional environments defined by the Main Board and ChiNext Board of China’s capital market. Drawing on Performance Feedback Theory and New Institutional Theory, it explores whether and how listing board heterogeneity moderates the relation between performance shortfalls and corporate strategic transformation, and further examines the boundary effects of internal governance factors including ownership type, CEO power, and resource slack.

Using panel data of A-share listed companies from 2009 to 2019 obtained from the CSMAR database, this study employs multiple regression analysis and interaction modeling to …

The Impact Of Value Co-Creation On Organizational Resilience In Small And Medium-Sized Traditional Foreign Trade Enterprises, Dan Yan

Dissertations and Theses Collection (Open Access)

Against the backdrop of increasing uncertainty in the global business environment, international scholarly attention to individual, group, and organizational resilience has risen significantly. Chinese export trading enterprises—especially small and medium-sized export trading enterprises(SMETEs)—face disadvantages such as small scale, weak financing capacity, limited access to information, and low brand strength. After experiencing shocks from digital transformation, the COVID-19 pandemic, the Russia–Ukraine war, and the European energy crisis, these firms have been compelled to reconsider how to enhance their organizational resilience in order to adapt to an increasingly turbulent external environment.

First, this study selects Company H as a representative case of …

Uniteformer: Unifying Node And Edge Modalities In Transformers For Vehicle Routing Problem, Dian Meng, Zhiguang Cao, Jie Gao, Yaoxin Wu, Yaqing Hou

Research Collection School Of Computing and Information Systems

Neural solvers for the Vehicle Routing Problem (VRP) have typically relied on either node or edge inputs, limiting their flexibility and generalization in real-world scenarios. We propose UniteFormer, a unified neural solver that supports node-only, edge-only, and hybrid input types through a single model trained via joint edge-node modalities. UniteFormer introduces: (1) a mixed encoder that integrates graph convolutional networks and attention mechanisms to collaboratively process node and edge features, capturing cross-modal interactions between them; and (2) a parallel decoder enhanced with query mapping and a feed-forward layer for improved representation. The model is trained with REINFORCE by randomly sampling …

Learning Memory-Enhanced Improvement Heuristics For Flexible Job Shop Scheduling, Jiaqi Wang, Zhiguang Cao, Peng Zhao, Rui Cao, Yubin Xiao, Yuan Jiang, You Zhou

Research Collection School Of Computing and Information Systems

The rise of smart manufacturing under Industry 4.0 introduces mass customization and dynamic production, demanding more advanced and flexible scheduling techniques. The flexible job-shop scheduling problem (FJSP) has attracted significant attention due to its complex constraints and strong alignment with real-world production scenarios. Current deep reinforcement learning (DRL)-based approaches to FJSP predominantly employ constructive methods. While effective, they often fall short of reaching (near-)optimal solutions. In contrast, improvement-based methods iteratively explore the neighborhood of initial solutions and are more effective in approaching optimality. However, the flexible machine allocation in FJSP poses significant challenges to the application of this framework, including …

Instance-Level Video Depth In Groups Beyond Occlusions, Yuan Liang, Yang Zhou, Ziming Sun, Tianyi Xiang, Guiqing Li, Shengfeng He

Research Collection School Of Computing and Information Systems

Depth estimation in dynamic, multi-object scenes remains a major challenge, especially under severe occlusions. Existing monocular models, including foundation models, struggle with instance-wise depth consistency due to their reliance on global regression. We tackle this problem from two key aspects: data and methodology. First, we introduce the Group Instance Depth (GID) dataset, the first large-scale video depth dataset with instance-level annotations, featuring 101,500 frames from real-world activity scenes. GID bridges the gap between synthetic and real-world depth data by providing high-fidelity depth supervision for multi-object interactions. Second, we propose InstanceDepth, the first occlusion-aware depth estimation framework for multi-object environments. Our …

Robust Hallucination Detection In Llms Via Adaptive Token Selection, Mengjia Niu, Hamed Haddadi, Guansong Pang

Research Collection School Of Computing and Information Systems

Hallucinations in large language models (LLMs) pose significant safety concerns that impede their broader deployment. Recent research in hallucination detection has demonstrated that LLMs’ internal representations contain truthfulness hints, which can be harnessed for detector training. However, the performance of these detectors is heavily dependent on the internal representations of predetermined tokens, fluctuating considerably when working on free-form generations with varying lengths and sparse distributions of hallucinated entities. To address this, we propose HaMI, a novel approach that enables robust detection of hallucinations through adaptive selection and learning of critical tokens that are most indicative of hallucinations. We achieve this …

Sempo: Lightweight Foundation Models For Time Series Forecasting, Hui He, Kun Yi, Yuanchi Ma, Qi Zhang, Zhengdong Niu, Guansong Pang

Research Collection School Of Computing and Information Systems

The recent boom of large pre-trained models witnesses remarkable success in developing foundation models (FMs) for time series forecasting. Despite impressive performance across diverse downstream forecasting tasks, existing time series FMs possess massive network architectures and require substantial pre-training on large-scale datasets, which significantly hinders their deployment in resource-constrained environments. In response to this growing tension between versatility and affordability, we propose SEMPO, a novel lightweight foundation model that requires pretraining on relatively small-scale data, yet exhibits strong general time series forecasting. Concretely, SEMPO comprises two key modules: 1) energy-aware SpEctral decomposition module, that substantially improves the utilization of pre-training …

General Test-Time Backdoor Detection In Split Neural Network-Based Vertical Federated Learning, Shunjie Yuan, Xinghua Li, Xuelin Cao, Haiyan Zhang, Robert H. Deng

Research Collection School Of Computing and Information Systems

As a new distributed machine learning framework, vertical federated learning (VFL) has been widely applied in the industry. However, recent studies have demonstrated that VFL faces serious challenges from backdoor attacks, which significantly hinder its further development. Although a few studies have focused on defending against VFL backdoor attacks, these defenses either do not consider the latest attack methods or show limited effectiveness. Moreover, most existing backdoor defense efforts primarily focus on backdoor attacks in horizontal federated learning (HFL) and centralized learning. Due to the unique architecture of VFL models, these methods cannot be directly applied to backdoor defense in …

From Discrete Manufacturing To Continuous Manufacturing: Examining The Relationship Among Digital Capability, Organizational Learning, And Enterprise Performance, Demu Chen

Dissertations and Theses Collection (Open Access)

In 2018, Company J launched its intelligent drive digital transformation project, which was completed and entered operation in 2021. That December, the project obtained certification under Zhejiang Province's "1353" system for the future factory enterprises, marking a successful transition from discrete to continuous manufacturing. To address the asynchronous flows of logistics, information, personnel, capital, and value indiscrete manufacturing enterprises, as well as pain points such as lowper capita output, long product delivery cycles, and low annual inventory turnover rates, this study, based on organizational learning theory, constructs ananalytical model encompassing digital capability (independent variable X), organizational learning (mediating variable Z), …

Riding The Waves Of Power: Power Fluctuation, Cognitive Energy, And Goal Pursuit, Hae-Lueng Rose Kim, Trevor A. Foulk, Michael Schaerer, Jake Gale, Eric A. Anicich

Research Collection Lee Kong Chian School Of Business

A central finding in the power literature is that experiencing elevated power facilitates employees’ goal-relevant cognitions and behaviors. In this work, we suggest that the relationship between power and goal pursuit is more complex than previously assumed. Specifically, we examine how experiencing power fluctuation—alternating states of high and low power during the workday—can uniquely promote employees’ goal-relevant behaviors beyond the effect of static power. Integrating insights from the Dynamic Equilibrium Model of Organizing (DEMO) and the Model of Proactive Motivation (MPM), our work demonstrates that power fluctuation can facilitate employees’ cognitive energy, in a way that enhances their goal-relevant cognitions …

Over-Reliance On Aesthetics? The Appearance-Reveals-Character Lay Theory Increases Consumers’ Devaluation Of Unattractive Produce, Shilpa Madan, Krishna Savani, Gita Venkataramani Johar

Research Collection Lee Kong Chian School Of Business

Approximately 40% of all produce is rejected because it appears unattractive, contributing significantly to food waste. Whereas previous research attributes this devaluation to the ugly-is-bad effect, this research identifies an important moderator of consumers’ reliance on this heuristic: the lay theory that a person’s outer appearance reveals their inner character. Specifically, consumers who believe that people’s appearances reveal their character are less willing to accept unattractive (vs. attractive) produce because they generalize their lay theory to produce and are, thus, more likely to infer that unattractive produce is of lower quality. Consumers who do not hold the lay theory do …

Growing Up Under Mao And Deng: On Politician Ideology And Corporate Policies, Hao Liang, Rong Wang, Haikun Zhu

Research Collection Lee Kong Chian School Of Business

We provide firm-level evidence on how politicians’ ideologies shape economic outcomes, using a unique setting of ideological discontinuity in China, transitioning from Maoism to Dengism around 1978. We find that ideological exposure during a politician’s early adulthood has a lasting impact on contemporary firm policies in their city. Firms governed by “Maoist mayors” show greater stakeholder spending, lower pay inequality, and less internationalization compared with those governed by “Dengist mayors.” Further evidence suggests that politicians’ ideologies influence corporate behaviors through nonpolicy channels, and these results cannot be fully explained by selection bias, endogenous matching, or the age of the mayor.

Framerate Sensitivity And Cinemagoing Explain The Soap Opera Effect Of Films: A Preregistered Study Of Undergraduate Students In Singapore., Sonny Rosenthal, Benjamin J. Li

Research Collection College of Integrative Studies

The soap opera effect is an unsettling feeling that some individuals experience while watching films at a high framerate. It has received little scholarly attention, and most explanations of it are speculative or anecdotal. Drawing on the mere exposure effect, this preregistered laboratory experiment provides new evidence of framerate sensitivity and cinemagoing as explanatory factors of the soap opera effect. Undergraduate students (N = 270) reported their cinemagoing and completed a novel task to measure their framerate sensitivity. They also completed a task indicating their framerate preferences. Those with a higher framerate sensitivity and more regular cinemagoing preferred the standard …

Registration Is A Powerful Rotation-Invariance Learner For 3d Anomaly Detection, Yuyang Yu, Zhengwei Chen, Xuemiao Xu, Lei Zhang, Haoxin Yang, Yongwei Nie, Shengfeng He

Research Collection School Of Computing and Information Systems

3D anomaly detection in point-cloud data is critical for industrial quality control, aiming to identify structural defects with high reliability. However, current memory bank-based methods often suffer from inconsistent feature transformations and limited discriminative capacity, particularly in capturing local geometric details and achieving rotation invariance. These limitations become more pronounced when registration fails, leading to unreliable detection results. We argue that point-cloud registration plays an essential role not only in aligning geometric structures but also in guiding feature extraction toward rotation-invariant and locally discriminative representations. To this end, we propose a registration-induced, rotation-invariant feature extraction framework that integrates the objectives …

Safe-Sora: Safe Text-To-Video Generation Via Graphical Watermarking, Zihan Su, Xuerui Qiu, Hongbin Xu, Tangyu Jiang, Jun-Hao Zhuang, Chun Yuan, Ming Li, Shengfeng He, Fei Yu

Research Collection School Of Computing and Information Systems

The explosive growth of generative video models has amplified the demand for reliable copyright preservation of AI-generated content. Despite its popularity in image synthesis, invisible generative watermarking remains largely underexplored in video generation. To address this gap, we propose Safe-Sora, the first framework to embed graphical watermarks directly into the video generation process. Motivated by the observation that watermarking performance is closely tied to the visual similarity between the watermark and cover content, we introduce a hierarchical coarse-to-fine adaptive matching mechanism. Specifically, the watermark image is divided into patches, each assigned to the most visually similar video frame, and further …

Stableguard: Towards Unified Copyright Protection And Tamper Localization In Latent Diffusion Models, Haoxin Yang, Bangzhen Liu, Xuemiao Xu, Cheng Xu, Yuyang Yu, Zikai Huang, Yi Wang, Shengfeng He

Research Collection School Of Computing and Information Systems

The advancement of diffusion models has enhanced the realism of AI-generated content but also raised concerns about misuse, necessitating robust copyright protection and tampering localization. Although recent methods have made progress toward unified solutions, their reliance on post hoc processing introduces considerable application inconvenience and compromises forensic reliability. We propose StableGuard, a novel framework that seamlessly integrates a binary watermark into the diffusion generation process, ensuring copyright protection and tampering localization in Latent Diffusion Models through an end-to-end design. We develop a Multiplexing Watermark VAE (MPW-VAE) by equipping a pretrained Variational Autoencoder (VAE) with a lightweight latent residual-based adapter, enabling …

Jury-And-Judge Chain-Of-Thought For Uncovering Toxic Data In 3d Visual Grounding, Kaixiang Huang, Qifeng Zhang, Jin Wang, Jingru Yang, Yang Zhou, Huan Yu, Guodong Lu, Shengfeng He

Research Collection School Of Computing and Information Systems

3D Visual Grounding (3DVG) faces persistent challenges due to coarse scene-level observations and logically inconsistent annotations, which introduce ambiguities that compromise data quality and hinder effective model supervision. To address these challenges, we introduce Refer-Judge, a novel framework that harnesses the reasoning capabilities of Multimodal Large Language Models (MLLMs) to identify and mitigate toxic data. At the core of Refer-Judge is a Jury-and-Judge Chain-of-Thought paradigm, inspired by the deliberative process of the judicial system. This framework targets the root causes of annotation noise: jurors collaboratively assess 3DVG samples from diverse perspectives, providing structured, multi-faceted evaluations. Judges then consolidate these insights …