Computer Sciences Commons^™

Unlocking The Planning Capabilities Of Llms Through Maximum Diversity Fine-Tuning, Wenjun Li, Changyu Chen, Pradeep Varakantham

Research Collection School Of Computing and Information Systems

Large language models (LLMs) have demonstrated impressive task-solving capabilities through prompting techniques and system designs, including solving planning tasks (e.g., math proofs, basic travel planning) when sufficient data is available online and used during pre-training. However, for planning tasks with limited prior data (e.g., blocks world, advanced travel planning), the performance of LLMs, including proprietary models like GPT and Gemini, is poor. This paper investigates the impact of fine-tuning on the planning capabilities of LLMs, revealing that LLMs can achieve strong performance in planning through substantial (tens of thousands of specific examples) fine-tuning. Yet, this process incurs high economic, time, …

Eduqate: Generating Adaptive Curricula Through Rmabs In Education Settings, Sidney Tio, Dexun Li, Pradeep Varakantham

Research Collection School Of Computing and Information Systems

There has been significant interest in the development of personalized and adaptive educational tools that cater to a student's individual learning progress. A crucial aspect in developing such tools is in exploring how mastery can be achieved across a diverse yet related range of content in an efficient manner. While Reinforcement Learning and Multi-armed Bandits have shown promise in educational settings, existing works often assume the independence of learning content, neglecting the prevalent interdependencies between such content. In response, we introduce Education Network Restless Multi-armed Bandits (EdNetRMABs), utilizing a network to represent the relationships between interdependent arms. Subsequently, we propose …

Towards Multimodal Empathetic Response Generation: A Rich Text-Speech-Vision Avatar-Based Benchmark, Han Zhang, Zixiang Meng, Meng Luo, Hong Han, Lizi Liao, Erik Cambria, Hao Fei

Research Collection School Of Computing and Information Systems

Empathetic Response Generation (ERG) is one of the key tasks of the affective computing area, which aims to produce emotionally nuanced and compassionate responses to user's queries. However, existing ERG research is predominantly confined to the singleton text modality, limiting its effectiveness since human emotions are inherently conveyed through multiple modalities. To combat this, we introduce an avatar-based Multimodal ERG (MERG) task, entailing rich text, speech, and facial vision information. We first present a large-scale high-quality benchmark dataset, AvaMERG, which extends traditional text ERG by incorporating authentic human speech audio and dynamic talking-face avatar videos, encompassing a diverse range of …

Exploring The Potential Of Large Language Models For Heterophilic Graphs, Yuxia Wu, Shujie Li, Yuan Fang, Chuan Shi

Research Collection School Of Computing and Information Systems

Large language models (LLMs) have presented significant opportunities to enhance various machine learning applications, including graph neural networks (GNNs). By leveraging the vast open-world knowledge within LLMs, we can more effectively interpret and utilize textual data to better characterize heterophilic graphs, where neighboring nodes often have different labels. However, existing approaches for heterophilic graphs overlook the rich textual data associated with nodes, which could unlock deeper insights into their heterophilic contexts. In this work, we explore the potential of LLMs for modeling heterophilic graphs and propose a novel two-stage framework: LLM-enhanced edge discriminator and LLM-guided edge reweighting. In the first …

On Learning Informative Trajectory Embeddings For Imitation, Classification And Regression, Zichang Ge, Changyu Chen, Arunesh Sinha, Pradeep Varakantham

Research Collection School Of Computing and Information Systems

In real-world sequential decision making tasks like autonomousdriving, robotics, and healthcare, learning from observed state-action trajectories is critical for tasks like imitation, classification,and clustering. For example, self-driving cars must replicate humandriving behaviors, while robots and healthcare systems benefitfrom modeling decision sequences, whether or not they come fromexpert data. Existing trajectory encoding methods often focus onspecific tasks or rely on reward signals, limiting their ability togeneralize across domains and tasks.Inspired by the success of embedding models like CLIP andBERT in static domains, we propose a novel method for embeddingstate-action trajectories into a latent space that captures the skillsand competencies in the …

“Ronaldo’S A Poser!”: How The Use Of Generative Ai Shapes Debates In Online Forums, Yuhan Zeng, Yingxuan Shi, Xuehan Huang, Fiona Fui-Hoon Nah, Ray Lc

Research Collection School Of Computing and Information Systems

Online debates can enhance critical thinking but may escalate into hostile attacks. As humans are increasingly reliant on Generative AI (GenAI) in writing tasks, we need to understand how people utilize GenAI in online debates. To examine the patterns of writing behavior while making arguments with GenAI, we created an online forum for soccer fans to engage in turn-based and free debates in a post format with the assistance of ChatGPT, arguing on the topic of "Messi vs Ronaldo". After 13 sessions of two-part study and semi-structured interviews with 39 participants, we conducted content and thematic analyses to integrate insights …

Augsso: Secure Threshold Single-Sign-On Authentication With Popular Password Collection, Changsong Jiang, Chunxiang Xu, Guomin Yang

Research Collection School Of Computing and Information Systems

Single-sign-on authentication is widely deployed in mobile systems, which allows an identity server to authenticate a mobile user and issue her/him with a token, such that the user can access diverse mobile services. To address the single-point-offailure problem, threshold single-sign-on authentication (PbTA) is a feasible solution, where multiple identity servers perform user authentication and token issuance in a threshold way. However, existing PbTA schemes confront critical drawbacks. Specifically, these schemes are vulnerable to perpetual secret leakage attacks (PSLA): an adversary perpetually compromises secrets of identity servers (e.g., secret key shares or credentials) to break security. Besides, they fail to achieve …

Hdwsa2: A Secure Hierarchical Deterministic Wallet Supporting Stealth Address And Signature Aggregation, Xin Yin, Zhen Liu, Guomin Yang, Guoxing Chen, Haojin Zhu

Research Collection School Of Computing and Information Systems

Hierarchical Deterministic Wallet (HDW) and Stealth Address (SA) are widely used in cryptocurrency communities due to their functionality and security. In the preliminary version of this work (ESORICS 2022), we formally define the syntax and security models of Hierarchical Deterministic Wallet supporting Stealth Address (HDWSA), capturing the functionality and security requirements imposed by the practice in cryptocurrency. We propose a concrete HDWSA construction and prove its security in the random oracle model. Note that when applied in blockchain, in practice, signature aggregation could reduce the cost of computation, storage, and communication dramatically. In this full version, we develop HDWSA definition …

Reverse Modeling In Large Language Models, Sicheng Yu, Yuanchen Xu, Cunxiao Du, Yanying Zhou, Minghui Qiu, Qianru Sun, Hao Zhang, Jiawei Wu

Research Collection School Of Computing and Information Systems

Humans are accustomed to reading and writing in a forward manner, and this natural bias extends to text understanding in auto-regressive large language models (LLMs). This paper investigates whether LLMs, like humans, struggle with reverse modeling, specifically with reversed text inputs. We found that publicly available pre-trained LLMs cannot understand such inputs. However, LLMs trained from scratch with both forward and reverse texts can understand them equally well during inference. Our case study shows that different-content texts result in different losses if input (to LLMs) in different directions---some get lower losses for forward while some for reverse. This leads us …

Grounding Ai Use In Learning Science: A Conversation With Steven Miller, Steven Miller, Lieven Demeester

CASTLe: Collection of Articles on Scholarship for Teaching and Learning

In this insightful interview, SMU Associate Provost (Teaching and Learning Innovation) Lieven Demeester and Professor Emeritus of Information Systems Steven Miller discuss the integration of artificial intelligence (AI) in teaching and learning, emphasising the importance of grounding AI use in the fundamentals of learning science. They explore the evolving role of education in the context of AI advancements, highlighting the need for educators to focus on the cognitive aspects of learning, such as goal-directed practice and feedback. They also address the potential of AI as a collaborative agent in group projects and the importance of maintaining accountability and quality control …

Network-Based Crypto Asset Analysis, Ling Cheng

Dissertations and Theses Collection (Open Access)

The rise of cryptocurrency, particularly Bitcoin (BTC), has revolutionized the financial landscape, enabling decentralized, peer-to-peer transactions without the need for intermediaries such as banks or financial institutions. Since its inception in 2009, Bitcoin has grown exponentially, not only in terms of market value but also in its impact on global finance. However, together with this popularity comes a wide range of cybercrimes including hacking, Ponzi schemes, wash trading, extortion, and money laundering. As noted in recent research, the volume of illicit cryptocurrency activities has grown significantly, with billions of dollars in crypto assets being stolen or used for illegal purposes …

Towards Testing, Detecting, And Debloating Insecure Components In Android Applications, Zicheng Zhang

Dissertations and Theses Collection (Open Access)

The Android ecosystem’s openness and extensibility have fueled its dominance in the mobile market, but they also broaden the attack surface of applications by introducing insecure or redundant methods. Vulnerabilities arise from various sources, including insecure API usage, code cloning, and feature bloat, especially from unneeded components introduced during development. To address these challenges, this dissertation presents a systematic, three-phase pipeline that transitions seamlessly from vulnerability discovery to clone-based detection and, ultimately, to dynamic mitigation through runtime debloating. Each phase builds upon the insights and limitations of the previous, collectively forming a practical approach to improving Android app security.

In …

Ivyapc: Auditable Generalized Payment Channels, Ming Li, Yuxian Li, Jian Weng, Yingjiu Li, Jiasi Weng, Junzuo Lai, Robert H. Deng

Research Collection School Of Computing and Information Systems

Payment channels are a cornerstone of a scalable blockchain infrastructure that enables transacting parties to lock assets on the blockchain and perform rapid off-chain updates with minimal latency and overhead. These protocols dramatically reduce on-chain interaction and improve throughput, with blockchain consensus only invoked in the event of disputes or final closure. While widely adopted in single-chain settings—such as in the Lightning Network for Bitcoin—existing constructions have several limitations, in particular they suffer from at least one of the following limitations: 1. No cross-chain. They do not enable fast trading of assets that reside on multiple isolated blockchains. 2. Non-optimal …

How Developers Interact With Ai: A Taxonomy Of Human-Ai Collaboration In Software Engineering, Christoph Treude, Marco A. Gerosa

Research Collection School Of Computing and Information Systems

Artificial intelligence (AI), including large language models and generative AI, is emerging as a significant force in software development, offering developers powerful tools that span the entire development lifecycle. Although software engineering research has extensively studied AI tools in software development, the specific types of interactions between developers and these AI-powered tools have only recently begun to receive attention. Understanding and improving these interactions has the potential to enhance productivity, trust, and efficiency in AI-driven workflows. In this paper, we propose a taxonomy of interaction types between developers and AI tools, identifying eleven distinct interaction types, such as auto-complete code …

Bigcodebench: Benchmarking Code Generation With Diverse Function Calls And Complex Instructions, T.Y. Zhuo, M.C. Vu, J. Chim, ..., David Lo

Research Collection School Of Computing and Information Systems

Task automation has been greatly empowered by the recent advances in Large Language Models (LLMs) via Python code, where the tasks ranging from software engineering development to general-purpose reasoning. While current benchmarks have shown that LLMs can solve tasks using programs like human developers, the majority of their evaluations are limited to short and self-contained algorithmic tasks or standalone function calls. Solving challenging and practical tasks requires the capability of utilizing diverse function calls as tools to efficiently implement functionalities like data analysis and web development. In addition, using multiple tools to solve a task needs compositional reasoning by accurately …

One-For-All: Towards Universal Domain Translation With A Single Stylegan, Yong Du, Jiahui Zhan, Xinzhe Li, Junyu Dong, Sheng Chen, Ming-Hsuan Yang, Shengfeng He

Research Collection School Of Computing and Information Systems

In this paper, we propose a novel translation model, UniTranslator, for transforming representations between visually distinct domains under conditions of limited training data and significant visual differences. The main idea behind our approach is leveraging the domain-neutral capabilities of CLIP as a bridging mechanism, while utilizing a separate module to extract abstract, domain-agnostic semantics from the embeddings of both the source and target realms. Fusing these abstract semantics with target-specific semantics results in a transformed embedding within the CLIP space. To bridge the gap between the disparate worlds of CLIP and StyleGAN, we introduce a new non-linear mapper, the CLIP2P …

Characterising Reproducibility Debt In Scientific Software: A Systematic Literature Review, Zara Hassan, Christoph Treude, Michael Norrish, Graham Williams, Alex Potanin

Research Collection School Of Computing and Information Systems

Context: In scientific software, the inability to reproduce results is often due to technical issues and challenges in recreating the full computational workflow from the original analysis. We conceptualise this problem as Reproducibility Debt (RpD). Much research has been performed to propose solutions to tackle these issues across various computational science disciplines. It is essential to identify and accumulate existing knowledge on reproducibility issues and state-of-the-art solutions so as to provide researchers and practitioners with information that enables further research activities and RpD management in practice. Objective: In the context of scientific software, we aim to characterise RpD by providing …

Pearl: Towards Permutation-Resilient Llms, Liang Chen, Li Shen, Yang Deng, Xiaoyan Zhao, Bin Liang, Kam-Fai Wong

Research Collection School Of Computing and Information Systems

The in-context learning (ICL) capability of large language models (LLMs) enables them to perform challenging tasks using provided demonstrations. However, ICL is highly sensitive to the ordering of demonstrations, leading to instability in predictions. This paper shows that this vulnerability can be exploited to design a natural attack - difficult for model providers to detect - that achieves nearly 80% success rate on LLaMA-3 by simply permuting the demonstrations. Existing mitigation methods primarily rely on post-processing and fail to enhance the model's inherent robustness to input permutations, raising concerns about safety and reliability of LLMs. To address this issue, we …

On Unraveling Student Resilience And Academic Performance In Higher Education, Aldy Gunawan, Ee-Peng Lim, Audrey Tedja Widjaja, William Tov, James Foo, Lieven Lode E. Demeester

Research Collection School Of Computing and Information Systems

The transition period from pre-tertiary to higher education levels is critical. We explore the role of resilience by conducting a survey to investigate students’ resilience and the relationship with overall academic performance, learning experience, and well-being. This effort is part of an initiative to develop strategies for better student engagement in the academic program, enhance their resilience, and prepare them for a competitive job market. We conclude that (i) high-resilience students are associated with better life satisfaction and are likely to perform well academically, (ii) a favorable learning environment supports students to study and perform well in the university, and …

Bootstrapping Language Models With Dpo Implicit Rewards, Changyu Chen, Zichen Liu, Chao Du, Tianyu Pang, Qian Liu, Arunesh Sinha, Pradeep Varakantham, Min Lin

Research Collection School Of Computing and Information Systems

Human alignment in large language models (LLMs) is an active area of research. A recent groundbreaking work, direct preference optimization (DPO), has greatly simplified the process from past work in reinforcement learning from human feedback (RLHF) by bypassing the reward learning stage in RLHF. DPO, after training, provides an implicit reward model. In this work, we make a novel observation that this implicit reward model can by itself be used in a bootstrapping fashion to further align the LLM. Our approach is to use the rewards from a current LLM model to construct a preference dataset, which is then used …

Prioritizing Speech Test Cases, Zhou Yang, Jieke Shi, Muhammad Hilmi Asyrofi, Bowen Xu, Xin Zhou, Donggyun Han, David Lo

Research Collection School Of Computing and Information Systems

As Automated Speech Recognition (ASR) systems gain widespread acceptance, there is a pressing need to rigorously test and enhance their performance. Nonetheless, the process of collecting and executing speech test cases is typically both costly and time-consuming. This presents a compelling case for the strategic prioritization of speech test cases, which consist of a piece of audio and the corresponding reference text. The central question we address is: In what sequence should speech test cases be collected and executed to identify the maximum number of errors at the earliest stage? In this study, we introduce PRiOritizing sPeecH tEsT …

Towards Real-World Unsupervised Anomaly Detection For Images, Zhonghang Liu

Dissertations and Theses Collection (Open Access)

In the era of big data, data quality plays a critical role in computer vision, where the reliability and purity of training images are essential for optimal performance. When training models such as image classifiers and object detectors, the quality of the training data directly influences the success of the model. In other words, if the training dataset is contaminated, the model’s performance might accordingly decrease.

To address this challenge, unsupervised anomaly detection (UAD) has become an attractive research area. By automatically removing these anomalous data points, UAD can help improve the accuracy and robustness of machine learning models in …

Temporal Relational Graph Convolutional Networks For Financial Applications, Brindha Priyadarshini Jeyaraman

Dissertations and Theses Collection (Open Access)

The financial industry operates within a highly dynamic and interconnected ecosystem, presenting unique challenges for predictive modeling and decision-making. Accurately forecasting financial performance, assessing credit risk, detecting fraud, and ensuring compliance require methodologies that can capture complex temporal, relational, and contextual dependencies within financial data. This thesis investigates the use of Temporal Relational Graph Convolutional Networks (TRGCNs) combined with financial knowledge graphs (FKGs) to address these challenges and enable advanced analytics in the financial domain. We introduce FintechKG, a financial knowledge graph constructed through a threedimensional information extraction process, incorporating entities, temporal dimensions, and domain-specific financial relationships. A TRGCN-based framework …

The Impact Of Ai Usage On Employee Work Outcomes: The Mediating Roles Of Personal Control And Job Insecurity And The Moderating Role Of Ai Trust, Tiantian Wang

Dissertations and Theses Collection (Open Access)

The widespread application of artificial intelligence (AI) technology in the workplace offers significant potential for process optimization andperformance improvement. However, the psychological mechanisms throughwhich AI usage affects employee outcomes remain underexplored. To address this gap, the present study investigated a sample of 170 employees froma media company in China, utilizing a three-wave longitudinal survey design. Specifically, this study examined how AI usage influenced employee creativity and task performance improvement through two mediatingmechanisms: the enhancement of personal control in problem-solving and the elicitation of job insecurity. Furthermore, the moderating role of trust in AI inthe relationship between AI usage and job …

What Is A Digital Twin Anyway? Deriving The Definition For The Built Environment From Over 15,000 Scientific Publications, Abdelrahman Mahmoud, Edgardo Macatulad, Binyu Lei, Matias Quintana, Clayton Miller, Filip Biljecki

Research Collection College of Integrative Studies

The concept of Digital Twins (DT) has attracted significant attention across various domains, particularly within the built environment. However, there is a sheer volume of definitions and the terminological consensus remains out of reach. The lack of a universally accepted definition leads to ambiguities in their conceptualization and implementation, and may cause miscommunication for both researchers and practitioners.We employed Natural Language Processing (NLP) techniques to systematically extract and analyze definitions of DTs from a corpus of more than 15,000 full-text articles spanning diverse disciplines. The study compares these findings with insights from an expert survey that included 52 experts. The …

Rethinking Light Decoder-Based Solvers For Vehicle Routing Problems, Ziwei Huang, Jianan Zhou, Zhiguang Cao, Yixin Xu

Research Collection School Of Computing and Information Systems

Light decoder-based solvers have gained popularity for solving vehicle routing problems (VRPs) due to their efficiency and ease of integration with reinforcement learning algorithms. However, they often struggle with generalization to larger problem instances or different VRP variants. This paper revisits light decoder-based approaches, analyzing the implications of their reliance on static embeddings and the inherent challenges that arise. Specifically, we demonstrate that in the light decoder paradigm, the encoder is implicitly tasked with capturing information for all potential decision scenarios during solution construction within a single set of embeddings, resulting in high information density. Furthermore, our empirical analysis reveals …

A Selective Vehicle Routing Problem For The Bloodmobile System, Aldy Gunawan, Samuel Alan Darmasaputra, Sy Hoang Do, Vincent F. Yu

Research Collection School Of Computing and Information Systems

Mobile blood collection has the advantage of greater reach compared to blood drives at fixed donation sites and is preferable for individuals with limited time or means of transportation. Bloodmobiles are widely used in healthcare logistics to increase the number of donors and donation frequency and to better match blood demand with collection. Bloodmobiles are stationed at predetermined locations, while shuttles are assigned to visit these locations to collect the donated blood. This problem is formulated as the Selective Vehicle Routing Problem under the Bloodmobile System (SVRP-BM). This research extends the Selective Vehicle Routing Problem with Integrated Tours problem (SVRPwIT) …

Can Llms Replace Manual Annotation Of Software Engineering Artifacts?, Toufique Ahmed, Premkumar Devanbu, Christoph Treude, Michael Pradel

Research Collection School Of Computing and Information Systems

Experimental evaluations of software engineering innovations, e.g., tools and processes, often include human-subject studies as a component of a multi-pronged strategy to obtain greater generalizability of the findings. However, human-subject studies in our field are challenging, due to the cost and difficulty of finding and employing suitable subjects, ideally, professional programmers with varying degrees of experience. Meanwhile, large language models (LLMs) have recently started to demonstrate human-level performance in several areas. This paper explores the possibility of substituting costly human subjects with much cheaper LLM queries in evaluations of code and code-related artifacts. We study this idea by applying six …

Dual Operation Aggregation Graph Neural Networks For Solving Flexible Job-Shop Scheduling Problem With Reinforcement Learning, Peng Zhao, You Zhou, Di Wang, Zhiguang Cao, Yubin Xiao, Xuan Wu, Yuanshu Li, Hongjia Liu, Wei Du, Yuan Jiang, Liupu Wang

Research Collection School Of Computing and Information Systems

With the widespread adoption of Internet Protocol (IP) communication technology and web-based platforms, cloud manufacturing has become a significant hallmark of Industry 4.0. Integrating graph algorithms into these web-enabled environments is crucial as they facilitate the representation and analysis of complex relationships in manufacturing processes, enabling efficient decision-making and adaptability in dynamic environments. As a key scheduling problem in cloud manufacturing, the flexible job-shop scheduling problem (FJSP) finds extensive applications in real-world scenarios. However, traditional FJSP-solving methods struggle to meet the efficiency and adaptability demands of cloud manufacturing due to generalization issues and excessive computational time, while reinforcement learning-based methods …

Comadice: Offline Cooperative Multi-Agent Reinforcement Learning With Stationary Distribution Shift Regularization, The Viet Bui, Tien Mai, Hong Thanh Nguyen

Research Collection School Of Computing and Information Systems

Offline reinforcement learning (RL) has garnered significant attention for its ability to learn effective policies from pre-collected datasets without the need for further environmental interactions. While promising results have been demonstrated in single-agent settings, offline multi-agent reinforcement learning (MARL) presents additional challenges due to the large joint state-action space and the complexity of multi-agent behaviors. A key issue in offline RL is the distributional shift, which arises when the target policy being optimized deviates from the behavior policy that generated the data. This problem is exacerbated in MARL due to the interdependence between agents' local policies and the expansive joint …