Open Access. Powered by Scholars. Published by Universities.®
- Discipline
-
- Physical Sciences and Mathematics (9338)
- Social and Behavioral Sciences (9166)
- Computer Sciences (9030)
- Business (6904)
- International and Area Studies (3679)
-
- Asian Studies (3587)
- Databases and Information Systems (3562)
- Economics (2446)
- Law (2305)
- Software Engineering (2215)
- Artificial Intelligence and Robotics (1906)
- Finance and Financial Management (1290)
- Public Affairs, Public Policy and Public Administration (1143)
- Information Security (1108)
- Sociology (1108)
- Engineering (1100)
- Psychology (1081)
- Organizational Behavior and Theory (1064)
- Numerical Analysis and Scientific Computing (1059)
- Graphics and Human Computer Interfaces (952)
- Accounting (916)
- Education (900)
- Econometrics (824)
- Technology and Innovation (817)
- Communication (811)
- Strategic Management Policy (732)
- Arts and Humanities (689)
- Corporate Finance (665)
- Political Science (665)
- Higher Education (598)
- Keyword
-
- Singapore (1056)
- China (354)
- Singapore Management University (275)
- Innovation (206)
- COVID-19 (177)
-
- Machine learning (175)
- Artificial intelligence (144)
- Social media (137)
- Deep learning (135)
- Sustainability (130)
- Higher education (129)
- Corporate governance (117)
- Culture (113)
- Asia (110)
- Technology (109)
- Trust (97)
- Pandemic (92)
- Creativity (85)
- Performance (84)
- Leadership (83)
- Well-being (82)
- Education (81)
- Climate change (77)
- Data mining (77)
- Privacy (74)
- Reinforcement learning (74)
- Collaboration (72)
- India (71)
- Marketing (71)
- Southeast Asia (70)
- Publication Year
- Publication
-
- Research Collection School Of Computing and Information Systems (8692)
- Research Collection Lee Kong Chian School Of Business (3273)
- Research Collection School of Social Sciences (2262)
- Research Collection Yong Pung How School Of Law (2076)
- Research Collection School Of Economics (1760)
-
- Dissertations and Theses Collection (Open Access) (891)
- Research Collection School Of Accountancy (815)
- Perspectives@SMU (663)
- Research Collection College of Integrative Studies (534)
- Knowledge@SMU (403)
- Asian Management Insights (317)
- SMU Press Releases and News (243)
- Research Collection Library (221)
- Social Space (210)
- Research Collection Institute of Service Excellence (2007-2024) (85)
- Sim Kee Boon Institute for Financial Economics (62)
- Oral History Collection (50)
- Singapore Law Journal (Lexicon) (50)
- Digital Narratives of Asia (46)
- Dissertations and Theses Collection (45)
- Research@SMU: Connecting the Dots (45)
- Research Collection BNP Paribas Hedge Fund Centre (37)
- ROSA Research Briefs (31)
- Report to Stakeholders (25)
- Lien Centre for Social Innovation: Research (22)
- Research Collection Lee Kong Chian School of Business (22)
- CCX Research (20)
- Friends of the Museum Collection (20)
- SMU Corporate Reports (19)
- Student Publications (18)
- Publication Type
- File Type
Articles 1321 - 1350 of 23193
Full-Text Articles in Entire DC Network
Why Does My Transaction Fail? A First Look At Failed Transactions On The Solana Blockchain, Xiaoye Zheng, Zhiyuan Wan, David Lo, Difan Xie, Xiaohu Yang
Why Does My Transaction Fail? A First Look At Failed Transactions On The Solana Blockchain, Xiaoye Zheng, Zhiyuan Wan, David Lo, Difan Xie, Xiaohu Yang
Research Collection School Of Computing and Information Systems
Solana is an emerging blockchain platform, recognized for its high throughput and low transaction costs, positioning it as a preferred infrastructure for Decentralized Finance (DeFi), Non-Fungible Tokens (NFTs), and other Web 3.0 applications. In the Solana ecosystem, transaction initiators submit various instructions to interact with a diverse range of Solana smart contracts, among which are decentralized exchanges (DEXs) that utilize automated market makers (AMMs), allowing users to trade cryptocurrencies directly on the blockchain without the need for intermediaries. Despite the high throughput and low transaction costs of Solana, the advantages have exposed Solana to bot spamming for financial exploitation, resulting …
Less Is More: On The Importance Of Data Quality For Unit Test Generation, Junwei Zhang, Xing Hu, Shan Gao, Xin Xia, David Lo, Shanping Li
Less Is More: On The Importance Of Data Quality For Unit Test Generation, Junwei Zhang, Xing Hu, Shan Gao, Xin Xia, David Lo, Shanping Li
Research Collection School Of Computing and Information Systems
Unit testing is crucial for software development and maintenance. Effective unit testing ensures and improves software quality, but writing unit tests is time-consuming and labor-intensive. Recent studies have proposed deep learning (DL) techniques or large language models (LLMs) to automate unit test generation. These models are usually trained or fine-tuned on large-scale datasets. Despite growing awareness of the importance of data quality, there has been limited research on the quality of datasets used for test generation. To bridge this gap, we systematically examine the impact of noise on the performance of learning-based test generation models. We first apply the open …
Adapting Large Language Models For Parameter-Efficient Log Anomaly Detection, Ying Fu Lim, Jiawen Zhu, Guansong Pang
Adapting Large Language Models For Parameter-Efficient Log Anomaly Detection, Ying Fu Lim, Jiawen Zhu, Guansong Pang
Research Collection School Of Computing and Information Systems
Log Anomaly Detection (LAD) seeks to identify atypical patterns in log data that are crucial to assessing the security and condition of systems. Although Large Language Models (LLMs) have shown tremendous success in various fields, the use of LLMs in enabling the detection of log anomalies is largely unexplored. This work aims to fill this gap. Due to the prohibitive costs involved in fully fine-tuning LLMs,we explore the use of parameter-efficient fine-tuning techniques (PEFTs) for adapting LLMs to LAD.To have an in-depth exploration of the potential of LLM-driven LAD, we present a comprehensive investigation of leveraging two of the most …
A Rating System To Evaluate Non-Gaap Exclusion Quality, Patricia M. Dechow, Wei Ting Loh, Annika Yu Wang
A Rating System To Evaluate Non-Gaap Exclusion Quality, Patricia M. Dechow, Wei Ting Loh, Annika Yu Wang
Research Collection School Of Accountancy
We develop a rating system to evaluate the quality of individual non-GAAP exclusions. Our perspective is that high-quality exclusions reflect nonrecurring economic transactions, are transitory accounting adjustments, or have little usefulness in forecasting cash flows. We use four approaches to rate exclusions. We evaluate the serial correlation of the exclusion, survey accounting academics’ views, obtain practitioner ratings from the CFA Institute, and identify the exclusions approved by the Chinese securities regulator. A firm’s exclusion quality score is the weighted average rating of its individual exclusions. For our sample of S&P 500 firms, we document that exclusion quality varies by industry, …
Hardiness, Optimism And Hope: Longitudinal Study Of How Individuals May Build Resilience Through Experiential Undergraduate University Courses, Paul Lim, Chin Heng Low, Kevin Koh
Hardiness, Optimism And Hope: Longitudinal Study Of How Individuals May Build Resilience Through Experiential Undergraduate University Courses, Paul Lim, Chin Heng Low, Kevin Koh
Research Collection Lee Kong Chian School Of Business
This study explores the antecedents of students’ resilience in a management education setting, by looking at the role of state hope and dispositional optimism in cultivating their psychological hardiness. While resilience is recognized as integral to personal growth, extant research has predominantly focused on its correlations with psychological capital (PsyCap) elements such as hope and optimism; or examined the outcomes of hardiness as a proxy of resilience. Little has been done to explicate the directional causal relationships among the PsyCap elements and hardiness in view of resilience-building. Using a longitudinal mixed methods design, this study delves into how state hope, …
Lessons Learned From Sandboxing, Piloting And Policy Experimentation With Ai And Other Digital Initiatives: Part 2, Ten In-Depth Interviews, Steven M. Miller
Lessons Learned From Sandboxing, Piloting And Policy Experimentation With Ai And Other Digital Initiatives: Part 2, Ten In-Depth Interviews, Steven M. Miller
Research Collection School Of Computing and Information Systems
This report, "Lessons Learned from Sandboxing, Piloting and Policy Experimentation with AI and Other Digital Initiatives," captures insights and experiences from project experts involved in recent digital innovation initiatives with the governments of Bangladesh, Maldives, and Kazakhstan, and from project experts actively involved with the use of AI for delivering government digital services in the EU, New Zealand, Rwanda, Singapore, United States, and Uzbekistan. The ten in-depth interview write-ups produced from these nine different country settings provide a small but highly informative sample of rich descriptions of some of the important realities, approaches, nuances, issues and challenges related to testing …
Ntire 2025 Challenge On Event-Based Image Deblurring: Methods And Results, Lei Sun, Et. Al.
Ntire 2025 Challenge On Event-Based Image Deblurring: Methods And Results, Lei Sun, Et. Al.
Research Collection School Of Computing and Information Systems
This paper presents an overview of NTIRE 2025, the First Challenge on Event-Based Image Deblurring, detailing the proposed methodologies and corresponding results. The primary goal of the challenge is to design an event-based method that achieves high-quality image deblurring, with performance quantitatively assessed using Peak Signal-toNoise Ratio (PSNR). Notably, there are no restrictions on computational complexity or model size. The task focuses on leveraging both events and images as inputs for singleimage deblurring. A total of 199 participants registered, among whom 15 teams successfully submitted valid results, offering valuable insights into the current state of eventbased image deblurring. We anticipate …
Group-And-Match Vs. Route-Then-Insert: Order Dispatching In Vehicle-Based Dual Services (Vedus), Yue Lin, Hai Yang, Hai Wang
Group-And-Match Vs. Route-Then-Insert: Order Dispatching In Vehicle-Based Dual Services (Vedus), Yue Lin, Hai Yang, Hai Wang
Research Collection School Of Computing and Information Systems
Rapid urban transportation and delivery demand and relevant resource constraints have driven the need for more efficient vehicle utilization. An innovative concept, “Vehicle-based MultiServices” (VeMuS), is a service model in which a single vehicle offers multiple services simultaneously in an urban mobility system. Similarly, “Vehicle-based Dual Services” (VeDuS) refers to a vehicle that provides two services simultaneously (Sun et al., 2023).
Individual Or Team Analyst Reports? The Organization Of Analyst Research Activities, Xia Chen, Ning Jia, Dan Wang
Individual Or Team Analyst Reports? The Organization Of Analyst Research Activities, Xia Chen, Ning Jia, Dan Wang
Research Collection School Of Accountancy
Given the importance of research resource allocation within brokerage firms, we examine key factors that influence the issuance of individual vs. team analyst reports. Using a comprehensive sample of analyst reports from China for the 2008-2021 period, we find that this decision is influenced by (1) the brokerage firm's client interests, whereby firms held by the brokerage firm's mutual fund clients and firms that are the brokerage firm's underwriting clients receive more team than individual reports from the brokerage firm, and (2) the nature of corporate events, whereby routine events receive more team reports and nonroutine events receive more individual …
Bogging Down Investors: An Unintended Consequence Of Litigation Risk, Siwen Fu, Ke Wang, Liandong Zhang, Liu Zheng
Bogging Down Investors: An Unintended Consequence Of Litigation Risk, Siwen Fu, Ke Wang, Liandong Zhang, Liu Zheng
Research Collection School Of Accountancy
Securities litigation risk is a well-recognized yet underexplored source of financial reporting complexity or unreadability. This study examines the effect of litigation risk on the readability of corporate financial reports. The 1999 Silicon Graphics Inc. (SGI) court ruling unexpectedly reduced litigation risk for firms within the Ninth Circuit Court's jurisdiction. Using a difference-in-differences design centered on the SGI court ruling, we find that, while the readability of financial reports generally declines over the sample period, treated firms in the Ninth Circuit experience a comparatively smaller decline in readability than control firms in other states after the ruling. Put differently, treated …
How Do Institutional Investors Facilitate Reporting Comparability? Evidence From Common Institutional Ownership In The United States, Xuanbo Li, Yun Lou, Rencheng Wang, Kaitang Zhou
How Do Institutional Investors Facilitate Reporting Comparability? Evidence From Common Institutional Ownership In The United States, Xuanbo Li, Yun Lou, Rencheng Wang, Kaitang Zhou
Research Collection School Of Accountancy
We examine how common institutional investors (CIIs) facilitate the financial reporting comparability (FRC) of US firms. Common ownership increases FRC of firms that are directly owned by CIIs (via a direct effect) and has positive spillover effects on other firms in the same industry. We find spillover effects in two types of firms: (1) those that are commonly owned by different institutional investors but are connected through common firms, and (2) those that do not have any common ownership. These results suggest that the effect of common ownership goes beyond commonly owned firms and extends to non-commonly owned firms. Furthermore, …
The Experiential Value Of Gambling In F2p Gamespaces, Orlando Woods
The Experiential Value Of Gambling In F2p Gamespaces, Orlando Woods
Research Collection College of Integrative Studies
This article explores the experiential value of gambling within free-to-play (f2p) gamespaces. In doing so, it offers a more balanced understanding of the gambling/gaming interplay. Whilst existing understandings tend to focus on the assumedly predatory monetisation techniques that players are subjected to by game developers, I instead argue that a logic of gambling is integral to the f2p experience economy of Asian players. Drawing on an analysis of in-depth interviews with Singapore-based game players/developers, I show how the virtual items and worlds that players engage with through f2p gamespaces are imbued with experiential value, whilst developers leverage the volumes of …
Ecological Responsibility In The City: The Surfaces And Volumes Of Urban Mosquito Management, Orlando Woods, Sayd Randle
Ecological Responsibility In The City: The Surfaces And Volumes Of Urban Mosquito Management, Orlando Woods, Sayd Randle
Research Collection College of Integrative Studies
This paper explores how ecological responsibility becomes distributed across the surfaces and volumes of the city. Our focus is urban mosquito management – specifically, the management of the Aedes aegypti mosquito, the vector by which dengue and several other diseases are transmitted to humans – and the governance strategies deployed in Singapore to manage breeding sites and mosquito outbreaks. Whilst the governance of nature in Singapore is usually undertaken by the state, the effective management of mosquitoes requires responsibility for ecological management to become distributed throughout the city. Residents are expected to assume responsibility for surficial cleanliness and the removal …
Awareness, Knowledge, And Motivations About Lifespan, Healthspan, And Healthy Longevity Medicine In The General Population: The Healthy Longevity (Helo) Conceptual Framework, Belinda Wang, Anna Szucs, Elena Sandalova, E. J. Horberg, Paul A. O'Keefe, Louis Island, Hans J. Meij, Sonny Rosenthal, Andrea B. Maier
Awareness, Knowledge, And Motivations About Lifespan, Healthspan, And Healthy Longevity Medicine In The General Population: The Healthy Longevity (Helo) Conceptual Framework, Belinda Wang, Anna Szucs, Elena Sandalova, E. J. Horberg, Paul A. O'Keefe, Louis Island, Hans J. Meij, Sonny Rosenthal, Andrea B. Maier
Research Collection College of Integrative Studies
The global population is ageing and the gap between lifespan (total years lived) and healthspan (years lived free of diseases) is increasing. Healthy Longevity Medicine (HLM) is an approach to optimise health and healthspan, and it has substantial public health implications. Despite those implications, the understanding of public perspectives on this field is lacking. The HEalthy LOngevity (HELO) framework was developed through a literature review guided by expert discussions across disciplines to include evidence-based concepts of health-related decision-making, ageing, and HLM. The framework organises concepts into three components. The first two components, awareness and knowledge, explore public perception and understanding …
A Summoned Life: Vocation In The Analects, Teng-Kuan Ng
A Summoned Life: Vocation In The Analects, Teng-Kuan Ng
Research Collection College of Integrative Studies
This article explores the concept of vocation in the Analects. While most treatmentsof vocation focus on the Christian tradition, and while Confucian thought is oftenviewed as categorically nontheistic, Robert Merrihew Adams’s Finite and InfiniteGoods (1999) presents an illuminating comparative framework for tracing outKongzi’s religious ethical vision of how persons think, act, and feel when graspedby a sense of divine calling. I argue that in the Analects a vocation is compositelyconceived as: (1) a special divine imperative to an individual or a community (2) tolove particular goods within one’s actual circumstances (3) with nonconsequentialist,noncompetitive, and potentially counter-cultural commitment (4) such that …
Exposure To Different Motives Of Scientists Moderates Responses To Scientific Consensus: The Case Of Cultured Meat, Kimin Eom, Bryan Kwok Cheng Choy
Exposure To Different Motives Of Scientists Moderates Responses To Scientific Consensus: The Case Of Cultured Meat, Kimin Eom, Bryan Kwok Cheng Choy
Research Collection School of Social Sciences
This research examines how exposure to different motives of scientists affects the impact of scientific consensus on public attitudes toward cultured meat. While scientific consensus on the safety of cultured meat generally increased positive attitudes toward it, this effect depended on information about scientists’ motives. Exposure to information about scientists’ financial motives weakened the positive effect of scientific consensus because it undermined trust in scientists. Information about scientists’ prosocial motives did not influence the scientific consensus effect. These findings suggest that perceived motives can shift trust in experts, thereby affecting the influence of experts on public attitudes.
Potential And Pitfalls Of Romantic Artificial Intelligence (Ai) Companions: A Systematic Review, Qi Hui Jerlyn Ho, Meilan Hu, Tracy Xi Chen, Andree Hartanto
Potential And Pitfalls Of Romantic Artificial Intelligence (Ai) Companions: A Systematic Review, Qi Hui Jerlyn Ho, Meilan Hu, Tracy Xi Chen, Andree Hartanto
Research Collection School of Social Sciences
As Artificial Intelligence (AI) becomes more integrated into daily life, individuals have increasingly turned to AIdriven systems for emotional support, companionship, and even romantic relationships. These relationships can be both beneficial and detrimental. Given the need for a comprehensive understanding of this phenomenon, this systematic review uses Sternberg’s Triangular Theory of Love to provide a holistic summary of its key potentials and pitfalls. A total of 23 articles were identified from the following databases: EBSCOhost ERIC, EBSCOhost PsycInfo, PubMed, Scopus, and Web of Science. Results highlighted the key potentials of being in a romantic relationship with AI companions as: the …
Neuroticism Predicts Constructive Conflict Resolution Styles In Conscientious Individuals: Evidence From Self-Reports And Observer Ratings, Frosch Yi Xuan Quek, Hwajin Yang, Andree Hartanto
Neuroticism Predicts Constructive Conflict Resolution Styles In Conscientious Individuals: Evidence From Self-Reports And Observer Ratings, Frosch Yi Xuan Quek, Hwajin Yang, Andree Hartanto
Research Collection School of Social Sciences
The ability to resolve conflict is crucial for maintaining healthy relationships and psychological well-being. However, prior research has often overlooked the potential benefits of neuroticism for conflict resolution styles, due to a reliance on self-report measures and the lack of consideration of healthy neuroticism. This study examined the relationship between neuroticism and conflict resolution styles and highlights the pivotal role of conscientiousness and measurement modalities. Employing both self-reports and observer ratings of the Conflict Resolution Styles Inventory (CRSI), we analyzed data from young adults (N = 259) using structural equation modeling analysis. Self-reports indicated that neuroticism was associated with unconstructive …
Higher Education Is Under Siege: It Needs To Reinvent Itself, Lily Kong
Higher Education Is Under Siege: It Needs To Reinvent Itself, Lily Kong
Research Collection School of Social Sciences
In an op-ed for The Straits Times, SMU President Professor Lily Kong reflects on the urgent need for universities to respond to the growing convergence of pressures facing higher education – including political scrutiny, funding constraints, technological disruption, and public scepticism. She calls for a reimagining of the university’s purpose: one that moves beyond knowledge acquisition and employment outcomes to embrace holistic student development, lifelong learning, and social relevance. Highlighting the importance of resilience, self-awareness, and adaptability, Prof Kong emphasises that learning happens as much outside the classroom as within it – through mentoring, community engagement, reflection, and lived experience. …
‘Workism’ Isn’T The Enemy, Matthew Hammerton
‘Workism’ Isn’T The Enemy, Matthew Hammerton
Research Collection School of Social Sciences
In a commentary, SMU Associate Professor of Philosophy Matthew Hammerton opined that "workism" – finding meaning and identity in one’s job – should not be dismissed outright. He distinguished it from "statusism," which ties self-worth to social hierarchy and often leads to competition and misplaced values. In contrast, non-competitive workism can promote collaboration and meaningful engagement. While acknowledging concerns that workism may lead to imbalance or risk, he cautioned against pathologising the choice of those who derive deep fulfilment from their work.
Built Environment And Public Electric Vehicle Charging: An Investigation Using Poi Data And Computer Vision, Junfeng Jiao, Seung Jun Choi
Built Environment And Public Electric Vehicle Charging: An Investigation Using Poi Data And Computer Vision, Junfeng Jiao, Seung Jun Choi
Research Collection College of Integrative Studies
Public electric vehicle charging stations (EVCSs) are vital for boosting EV adoption. This study investigates Seoul’s public EV charging patterns, taking into account the surrounding urban built environment. We collected built-environment data from land-use maps, Point of Interest (POI) data, and panorama images near public EVCS. The computer-vision technique was used to extract scene features from panorama images. We conducted a spatiotemporal analysis of public EVCS usage. The built-environment factors underwent dimensionality reduction and were assessed for outliers. Descriptive analysis revealed afternoon peak charging times and variations between chargers. Additional peaks are observed in the weekday late evening for chargers …
Assessing Equity In Heat Mitigation Ecosystem Services Of Urban Green Space In Singapore, Ronghua Xu, Perrine Hamel, Youu Ming, Aikeen Lim, Tingting He
Assessing Equity In Heat Mitigation Ecosystem Services Of Urban Green Space In Singapore, Ronghua Xu, Perrine Hamel, Youu Ming, Aikeen Lim, Tingting He
Research Collection College of Integrative Studies
Urban green space provides ecosystem services such as heat mitigation, which are associated with improved resident health and welfare. While estimating the disparity in ecosystem services is essential to improve equity, little guidance is available as to which indices and indicators should be used. Here, we assess the distributional equity in heat mitigation benefits as measured by two indicators – land surface temperature reduction and wet-bulb temperature reduction – in the city-state of Singapore, and compare three inequality indices: Gini coefficient (GC), Atkinson index (AI), and Theil entropy index (TE). We include scenarios with and without migrant construction workers to …
On-Demand Heterogeneous Drone Delivery Problem, Xupeng Wen, Zhiguang Cao, Shu Xu, Dapeng Ren, Guohua Wu, Yaoxin Wu
On-Demand Heterogeneous Drone Delivery Problem, Xupeng Wen, Zhiguang Cao, Shu Xu, Dapeng Ren, Guohua Wu, Yaoxin Wu
Research Collection School Of Computing and Information Systems
In the on-demand problem domain, actual demand frequently deviates from the expected demand. This paper intricately delves into the exploration of on-demand heterogeneous multi-drone routing problem (ODHDRP), in which a transport drone carries multiple terminal drones to subregions in the first echelon, and the terminal drones deliver parcels during a flight trip to customers with demands in subregions to maintain economies of scale in the second echelon. We formulate the customer demands using a normal distribution, and exploit a reliability model of customer demands with chance constraints. To solve the ODHDRP efficiently, we propose a hybrid iterative optimisation heuristic (HIOH) …
Moditector: Module-Directed Testing For Autonomous Driving Systems, Renzhi Wang, Mingfei Cheng, Xiaofei Xie, Yuan Zhou, Lei Ma
Moditector: Module-Directed Testing For Autonomous Driving Systems, Renzhi Wang, Mingfei Cheng, Xiaofei Xie, Yuan Zhou, Lei Ma
Research Collection School Of Computing and Information Systems
Testing Autonomous Driving Systems (ADSs) is crucial for ensuring their safety, reliability, and performance. Despite numerous testing methods available that can generate diverse and challenging scenarios to uncover potential vulnerabilities, these methods often treat ADS as a black-box, primarily focusing on identifying system-level failures like collisions or near-misses without pinpointing the specific modules responsible for these failures. This lack of root causes understanding for the failures hinders effective debugging and subsequent system repair. Furthermore, current approaches often fall short in generating violations that adequately test the individual modules of an ADS from a system-level perspective, such as perception, prediction, planning, …
Demystifying Memorization In Llm-Based Program Repair Via A General Hypothesis Testing Framework, Jiaolong Kong, Xiaofei Xie, Shangqing Liu
Demystifying Memorization In Llm-Based Program Repair Via A General Hypothesis Testing Framework, Jiaolong Kong, Xiaofei Xie, Shangqing Liu
Research Collection School Of Computing and Information Systems
Large Language Models (LLMs) have achieved remarkable success in various applications, particularly in code-related tasks such as code generation and program repair, setting new performance benchmarks. However, the extensive use of large training corpora raises concerns about whether these achievements stem from genuine understanding or mere memorization of training data—a question often overlooked in current research. This paper aims to study the memorization issue within LLM-based program repair by investigating whether the correct patches generated by LLMs are the result of memorization. The key challenge lies in the absence of ground truth for confirming memorization, leading to various ad-hoc methods …
Collaborative Tree Search For Enhancing Embodied Multi-Agent Collaboration, Lizheng Zu, Lin Lin, Song Fu, Na Zhao, Pan Zhou
Collaborative Tree Search For Enhancing Embodied Multi-Agent Collaboration, Lizheng Zu, Lin Lin, Song Fu, Na Zhao, Pan Zhou
Research Collection School Of Computing and Information Systems
Embodied agents based on large language models (LLMs) face significant challenges in collaborative tasks, requiring effective communication and reasonable division of labor to ensure efficient and correct task completion. Previous approaches with simple communication patterns carry erroneous or incoherent agent actions, which can lead to additional risks. To address these problems, we propose Cooperative Tree Search (CoTS), a framework designed to significantly improve collaborative planning and task execution efficiency among embodied agents. CoTS guides multi-agents to discuss long-term strategic plans within a modified Monte Carlo tree, searching along LLMdriven reward functions to provide a more thoughtful and promising approach to …
Modfinity: Unsupervised Domain Adaptation With Multimodal Information Flow Intertwining, Shanglin Liu, Jianming Lv, Jingdan Kang, Huaidong Zhang, Zequan Liang, Shengfeng He
Modfinity: Unsupervised Domain Adaptation With Multimodal Information Flow Intertwining, Shanglin Liu, Jianming Lv, Jingdan Kang, Huaidong Zhang, Zequan Liang, Shengfeng He
Research Collection School Of Computing and Information Systems
Multimodal unsupervised domain adaptation leverages unlabeled data in the target domain to enhance multimodal systems continuously. While current state-of-the-art methods encourage interaction between sub-models of different modalities through pseudo-labeling and feature-level exchange, varying sample quality across modalities can lead to the propagation of inaccurate information, resulting in error accumulation. To address this, we propose Modal-Affinity Multimodal Domain Adaptation (MODfinity), a method that dynamically manages multimodal information flow through fine-grained control over teacher model selection, guiding information intertwining at both feature and label levels. By treating labels as an independent modality, MODfinity enables balanced performance assessment across modalities, employing a novel …
A Knowledge Enhanced Large Language Model For Bug Localization, Yue Li, Bohan Liu, Ting Zhang, Zhiqi Wang, David Lo, Lanxin Yang, Jun Lyu, He Zhang
A Knowledge Enhanced Large Language Model For Bug Localization, Yue Li, Bohan Liu, Ting Zhang, Zhiqi Wang, David Lo, Lanxin Yang, Jun Lyu, He Zhang
Research Collection School Of Computing and Information Systems
A significant number of bug reports are generated every day as software systems continue to develop. Large Language Models (LLMs) have been used to correlate bug reports with source code to locate bugs automatically. The existing research has shown that LLMs are effective for bug localization and can increase software development efficiency. However, these studies still have two limitations. First, these models fail to capture context information about bug reports and source code. Second, these models are unable to understand the domain-specific expertise inherent to particular projects, such as version information in projects that are composed of alphanumeric characters without …
Efficient And Green Large Language Models For Software Engineering: Literature Review, Vision, And The Road Ahead, Jieke Shi, Zhou Yang, David Lo
Efficient And Green Large Language Models For Software Engineering: Literature Review, Vision, And The Road Ahead, Jieke Shi, Zhou Yang, David Lo
Research Collection School Of Computing and Information Systems
Large Language Models (LLMs) have recently shown remarkable capabilities in various software engineering tasks, spurring the rapid growth of the Large Language Models for Software Engineering (LLM4SE) area. However, limited attention has been paid to developing efficient LLM4SE techniques that demand minimal computational cost, time, and memory resources, as well as green LLM4SE solutions that reduce energy consumption, water usage, and carbon emissions. This article aims to redirect the focus of the research community toward the efficiency and greenness of LLM4SE, while also sharing potential research directions to achieve this goal. It commences with a brief overview of the significance …
Deepvec: State-Vector Aware Test Case Selection For Enhancing Recurrent Neural Network, Zhonghao Jiang, Meng Yan, Li Huang, Weifeng Sun, Chao Liu, Song Sun, David Lo
Deepvec: State-Vector Aware Test Case Selection For Enhancing Recurrent Neural Network, Zhonghao Jiang, Meng Yan, Li Huang, Weifeng Sun, Chao Liu, Song Sun, David Lo
Research Collection School Of Computing and Information Systems
Deep Neural Networks (DNN) have realized significant achievements across various application domains. There is no doubt that testing and enhancing a pre-trained DNN that has been deployed in an application scenario is crucial, because it can reduce the failures of the DNN. DNN-driven software testing and enhancement require large amounts of labeled data. The high cost and inefficiency caused by the large volume of data of manual labeling, and the time consumption of testing all cases in real scenarios are unacceptable. Therefore, test case selection technologies are proposed to reduce the time cost by selecting and only labeling representative test …