Open Access. Powered by Scholars. Published by Universities.®
- Discipline
-
- Physical Sciences and Mathematics (8556)
- Computer Sciences (8539)
- Databases and Information Systems (3453)
- Software Engineering (2164)
- Artificial Intelligence and Robotics (1697)
-
- Information Security (1071)
- Numerical Analysis and Scientific Computing (1025)
- Graphics and Human Computer Interfaces (933)
- Engineering (891)
- Social and Behavioral Sciences (682)
- Business (646)
- Theory and Algorithms (493)
- Computer Engineering (459)
- Operations Research, Systems Engineering and Industrial Engineering (405)
- Programming Languages and Compilers (379)
- OS and Networks (324)
- Communication (300)
- Social Media (242)
- Public Affairs, Public Policy and Public Administration (212)
- Transportation (189)
- Medicine and Health Sciences (183)
- Education (172)
- Management Information Systems (168)
- Data Storage Systems (162)
- E-Commerce (148)
- International and Area Studies (114)
- Asian Studies (113)
- Health Information Technology (112)
- Digital Communications and Networking (109)
- Technology and Innovation (106)
- Keyword
-
- Deep learning (124)
- Machine learning (122)
- Social media (76)
- Artificial intelligence (71)
- Reinforcement learning (69)
-
- Privacy (65)
- Data mining (64)
- Cloud computing (58)
- Deep Learning (56)
- Empirical study (54)
- Security (54)
- Optimization (53)
- Visualization (52)
- Task analysis (50)
- Neural networks (49)
- Online learning (49)
- Singapore (49)
- Software engineering (49)
- Training (49)
- Anomaly detection (47)
- Feature extraction (46)
- Twitter (46)
- Large Language Models (45)
- Blockchain (44)
- Collaboration (44)
- Semantics (43)
- Access control (41)
- Algorithms (40)
- Android (39)
- Classification (38)
- Publication Year
- File Type
Articles 301 - 330 of 8612
Full-Text Articles in Entire DC Network
Griffin: Effective Token Alignment For Faster Speculative Decoding, Shijing Hu, Jingyang Li, Xingyu Xie, Zhihui Lu, Kim-Chuan Toh, Pan Zhou
Griffin: Effective Token Alignment For Faster Speculative Decoding, Shijing Hu, Jingyang Li, Xingyu Xie, Zhihui Lu, Kim-Chuan Toh, Pan Zhou
Research Collection School Of Computing and Information Systems
Speculative decoding accelerates inference in large language models (LLMs) by generating multiple draft tokens simultaneously. However, existing methods often struggle with token misalignment between the training and decoding phases, limiting their performance. To address this, we propose GRIFFIN, a novel framework that incorporates a token-alignable training strategy and a token-alignable draft model to mitigate misalignment. The training strategy employs a loss masking mechanism to exclude highly misaligned tokens during training, preventing them from negatively impacting the draft model’s optimization. The token-alignable draft model introduces input tokens to correct inconsistencies in generated features. Experiments on LLaMA, Vicuna, Qwen and Mixtral models …
Sopo: Text-To-Motion Generation Using Semi-Online Preference Optimization, Xiaofeng Tan, Hongsong Wang, Xin Geng, Pan Zhou
Sopo: Text-To-Motion Generation Using Semi-Online Preference Optimization, Xiaofeng Tan, Hongsong Wang, Xin Geng, Pan Zhou
Research Collection School Of Computing and Information Systems
Text-to-motion generation is essential for advancing the creative industry but often presents challenges in producing consistent, realistic motions. To address this, we focus on fine-tuning text-to-motion models to consistently favor highquality, human-preferred motions—a critical yet largely unexplored problem. In this work, we theoretically investigate the DPO under both online and offline settings, and reveal their respective limitation: overfitting in offline DPO, and biased sampling in online DPO. Building on our theoretical insights, we introduce Semi-online Preference Optimization (SoPo), a DPO-based method for training text-to-motion models using “semi-online” data pair, consisting of unpreferred motion from online distribution and preferred motion in …
Hybrid-Balance Gflownet For Solving Vehicle Routing Problems, Ni Zhang, Zhiguang Cao
Hybrid-Balance Gflownet For Solving Vehicle Routing Problems, Ni Zhang, Zhiguang Cao
Research Collection School Of Computing and Information Systems
Existing GFlowNet-based methods for vehicle routing problems (VRPs) typically employ Trajectory Balance (TB) to achieve global optimization but often neglect important aspects of local optimization. While Detailed Balance (DB) addresses local optimization more effectively, it alone falls short in solving VRPs, which inherently require holistic trajectory optimization. To address these limitations, we introduce the Hybrid-Balance GFlowNet (HBG) framework, which uniquely integrates TB and DB in a principled and adaptive manner by aligning their intrinsically complementary strengths. Additionally, we propose a specialized inference strategy for depot-centric scenarios like the Capacitated Vehicle Routing Problem (CVRP), leveraging the depot node's greater flexibility in …
Uniteformer: Unifying Node And Edge Modalities In Transformers For Vehicle Routing Problem, Dian Meng, Zhiguang Cao, Jie Gao, Yaoxin Wu, Yaqing Hou
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 …
Large Language Models As End-To-End Combinatorial Optimization Solvers, Xia Jiang, Yaoxin Wu, Minshuo Li, Zhiguang Cao, Yingqian Zhang
Large Language Models As End-To-End Combinatorial Optimization Solvers, Xia Jiang, Yaoxin Wu, Minshuo Li, Zhiguang Cao, Yingqian Zhang
Research Collection School Of Computing and Information Systems
Combinatorial optimization (CO) problems, central to decision-making scenarios like logistics and manufacturing, are traditionally solved using problem-specific algorithms requiring significant domain expertise. While large language models (LLMs) have shown promise in automating CO problem solving, existing approaches rely on intermediate steps such as code generation or solver invocation, limiting their generality and accessibility. This paper introduces a novel framework that empowers LLMs to serve as end-to-end CO solvers by directly mapping natural language problem descriptions to solutions. We propose a two-stage training strategy: supervised fine-tuning (SFT) imparts LLMs with solution generation patterns from domain-specific solvers, while a feasibility-and-optimality-aware reinforcement learning …
Multi-Task Vehicle Routing Solver Via Mixture Of Specialized Experts Under State-Decomposable Mdp, Yuxin Pan, Zhiguang Cao, Chengyang Gu, Liu Liu, Peilin Zhao, Yize Chen, Fangzhen Lin
Multi-Task Vehicle Routing Solver Via Mixture Of Specialized Experts Under State-Decomposable Mdp, Yuxin Pan, Zhiguang Cao, Chengyang Gu, Liu Liu, Peilin Zhao, Yize Chen, Fangzhen Lin
Research Collection School Of Computing and Information Systems
Existing neural methods for multi-task vehicle routing problems (VRPs) typically learn unified solvers to handle multiple constraints simultaneously. However, they often underutilize the compositional structure of VRP variants, each derivable from a common set of basis VRP variants. This critical oversight causes unified solvers to miss out the potential benefits of basis solvers, each specialized for a basis VRP variant. To overcome this limitation, we propose a framework that enables unified solvers to perceive the shared-component nature across VRP variants by proactively reusing basis solvers, while mitigating the exponential growth of trained neural solvers. Specifically, we introduce a State-Decomposable MDP …
Learning Memory-Enhanced Improvement Heuristics For Flexible Job Shop Scheduling, Jiaqi Wang, Zhiguang Cao, Peng Zhao, Rui Cao, Yubin Xiao, Yuan Jiang, You Zhou
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 …
Island-Based Evolutionary Computation With Diverse Surrogates And Adaptive Knowledge Transfer For High-Dimensional Data-Driven Optimization, Xianrong Zhang, Yuejiao Gong, Zhiguang Cao, Jun Zhang
Island-Based Evolutionary Computation With Diverse Surrogates And Adaptive Knowledge Transfer For High-Dimensional Data-Driven Optimization, Xianrong Zhang, Yuejiao Gong, Zhiguang Cao, Jun Zhang
Research Collection School Of Computing and Information Systems
In recent years, there has been a growing interest in data-driven evolutionary algorithms (DDEAs) employing surrogate models to approximate the objective functions with limited data. However, current DDEAs are primarily designed for lower-dimensional problems and their performance drops significantly when applied to large-scale optimization problems (LSOPs). To address the challenge, this paper proposes an offline DDEA named DSKT-DDEA. DSKT-DDEA leverages multiple islands that utilize different data to establish diverse surrogate models, fostering diverse subpopulations and mitigating the risk of premature convergence. In the intra-island optimization phase, a semi-supervised learning method is devised to fine-tune the surrogates. It not only facilitates …
Robust Hallucination Detection In Llms Via Adaptive Token Selection, Mengjia Niu, Hamed Haddadi, Guansong Pang
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 …
Semi‑Supervised Graph Anomaly Detection Via Robust Homophily Learning, Guoguo Ai, Hezhe Qiao, Hui Yan, Guansong Pang
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 …
Coresets For Clustering Under Stochastic Noise, Lingxiao Huang, Zhize Li, Nisheeth K. Vishnoi, Runkai Yang, Haoyu Zhao
Coresets For Clustering Under Stochastic Noise, Lingxiao Huang, Zhize Li, Nisheeth K. Vishnoi, Runkai Yang, Haoyu Zhao
Research Collection School Of Computing and Information Systems
We study the problem of constructing coresets for $(k, z)$-clustering when the input dataset is corrupted by stochastic noise drawn from a known distribution. In this setting, evaluating the quality of a coreset is inherently challenging, as the true underlying dataset is unobserved. To address this, we investigate coreset construction using surrogate error metrics that are tractable and provably related to the true clustering cost. We analyze a traditional metric from prior work and introduce a new error metric that more closely aligns with the true cost. Although our metric is defined independently of the noise distribution, it enables approximation …
Generalization Bounds For Rank‑Sparse Neural Networks, Antoine Ledent, Rodrigo Alves, Yunwen Lei
Generalization Bounds For Rank‑Sparse Neural Networks, Antoine Ledent, Rodrigo Alves, Yunwen Lei
Research Collection School Of Computing and Information Systems
It has been recently observed in much of the literature that neural networks exhibit a bottleneck rank property: for larger depths, the activation and weights of neural networks trained with gradient-based methods tend to be of approximately low rank. In fact, the rank of the activations of each layer converges to a fixed value referred to as the “bottleneck rank”, which is the minimum rank required to represent the training data. This perspective is in line with the observation that regularizing linear networks (without activations) with weight decay is equivalent to minimizing the Schatten p quasi norm of the neural …
Sempo: Lightweight Foundation Models For Time Series Forecasting, Hui He, Kun Yi, Yuanchi Ma, Qi Zhang, Zhengdong Niu, Guansong Pang
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 …
Design Principles For Customer‑Engaging Digital Service Systems: An Action Research Study, Keng Siau, Xiaofeng Chen, Xin Tan
Design Principles For Customer‑Engaging Digital Service Systems: An Action Research Study, Keng Siau, Xiaofeng Chen, Xin Tan
Research Collection School Of Computing and Information Systems
Digital services represent a business approach employed by organizations to operate in the digital environment. However, systematic development guidelines for developing quality digital service systems are lacking in the literature. The authors identified four general challenges for developing and implementing customer-engaging digital service systems (CEDSS). By employing the method of canonical action research in a digital service system project, they derived 10 design principles for developing high-quality CEDSS. They empirically evaluated the design principles in the development project and through follow-up focus group sessions. The design principles provide applicable and actionable guidelines for the development of CEDSS.
Accuracy-Enabling Differential Privacy-Preserving Truth Discovery, Man Zhang, Xinghua Li, Yinbin Miao, Bin Luo, Siqi Ma, Robert H. Deng
Accuracy-Enabling Differential Privacy-Preserving Truth Discovery, Man Zhang, Xinghua Li, Yinbin Miao, Bin Luo, Siqi Ma, Robert H. Deng
Research Collection School Of Computing and Information Systems
Perturbation-based privacy-preserving truth discovery requires the Service Provider (SP) to calculate the truthful aggregation result from perturbed data of the Data Sources (DSs), which inevitably damages the aggregation accuracy due to perturbation noise added in the data. Thus, the existing works attempt to relieve the perturbation errors by reducing noise amounts or adjusting aggregation weights of DSs. However, the former sacrifices DSs’ privacy preservation and the latter has the limited accuracy recovery performance. Aiming at it, we propose an accuracy-enabling differential privacy-preserving truth discovery consisting of an independence-guaranteed data perturbation module and a progressive-private noise elimination module. Specifically, in the …
Bias Testing And Mitigation In Llm-Based Code Generation, Dong Huang, Jie M. Zhang, Qingwen Bu, Xiaofei Xie, Junjie Chen, Heming Cui
Bias Testing And Mitigation In Llm-Based Code Generation, Dong Huang, Jie M. Zhang, Qingwen Bu, Xiaofei Xie, Junjie Chen, Heming Cui
Research Collection School Of Computing and Information Systems
As the adoption of LLMs becomes more widespread in software coding ecosystems, a pressing issue has emerged: does the generated code contain social bias and unfairness, such as those related to age, gender, and race? This issue concerns the integrity, fairness, and ethical foundation of software applications that depend on the code generated by these models but are underexplored in the literature. This paper presents a novel bias testing framework that is specifically designed for code generation tasks. Based on this framework, we conduct an extensive empirical study on the biases in code generated by five widely studied LLMs (i.e., …
A Rate-Dependent Coreset Selector For Continual Learning On Time-Varying Data Distributions, Zilin Luo, Zichen Tian, Yaoyao Liu, Qianru Sun
A Rate-Dependent Coreset Selector For Continual Learning On Time-Varying Data Distributions, Zilin Luo, Zichen Tian, Yaoyao Liu, Qianru Sun
Research Collection School Of Computing and Information Systems
In this paper we review the concept of “phase” defined in Class-Incremental Learning (CIL), i.e., learning new classes while not forgetting old ones. Due to this design, classic CIL algorithms are mostly offline or can handle only intensive data distribution shifts across the phases. However, real-world data streams are often online, usually with uncertain or untraceable changes in their data distributions. To this end, we design the per-step distribution shifts by modeling the class sampling weights using bell-shaped curves. Such a design respects the rise-and-fall nature and presents realistic but underexplored challenges for CIL: 1) The data non-stationarity across steps …
Genscore: Agent-Based Short-Answer Question Generation And Scoring In Software Engineering Courses, Nguyen Binh Duong Ta, Lwin Khin Shar
Genscore: Agent-Based Short-Answer Question Generation And Scoring In Software Engineering Courses, Nguyen Binh Duong Ta, Lwin Khin Shar
Research Collection School Of Computing and Information Systems
Short-answer questions are commonly used in educational assessments, as they are often viewed as a more effective way than multiple-choice questions to determine whether students have achieved the intended learning outcomes. However, manually creating appropriate questions targeting different cognitive levels such as those defined by the Bloom’s Taxonomy, and grading text answers from students are not trivial tasks for instructors. Existing work on auto-question generation and scoring in computing education typically targets coding-based questions. However, in software engineering courses, assessments can extend beyond coding to understanding of processes, DevOps methodologies, system design, etc. This work aims to address the dual …
Reliable-Data-Split (Rds): Maximizing Model Potential With Reinforced Selection Strategy, Hoang D. Nguyen, Xuan-Son Vu, Quoc Tuan Truong, Duc-Trong Le
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 …
Immersion In Virtual Cultural Heritage: Framework, Research Landscape, Future Directions, And Ongoing Study, Xuenan Huo, Keng Siau
Immersion In Virtual Cultural Heritage: Framework, Research Landscape, Future Directions, And Ongoing Study, Xuenan Huo, Keng Siau
Research Collection School Of Computing and Information Systems
This study proposes that a nuanced understanding of immersion’sdistinct dimensions and their mediating role between technical de-sign features and cultural learning outcomes is critical for advancingVirtual Cultural Heritage (VCH) research and practice. To advancethe immersion designs in the field of VCH, the study provides anoverview of the immersion nuances and research state-of-the-artof VCH studies. A tripartite framework of spatial, emotional, andsocial immersion in virtual environments is articulated. A litera-ture review was conducted based on this framework. The reviewreveals multiple research gaps and opportunities, including absenceof immersion, predominant focus on spatial immersion, lagged AIintegration, lack of accessibility and inclusive designs, and …
Mando-Llm: Heterogeneous Graph Transformers With Large Language Models For Smart Contract Vulnerability Detection, Nhat Minh Nguyen, Huu Hoang Nguyen, Long Le Thanh, Zahra Ahmadi, Thanh Nam Doan, Daoyuan Wu, Lingxiao Jiang
Mando-Llm: Heterogeneous Graph Transformers With Large Language Models For Smart Contract Vulnerability Detection, Nhat Minh Nguyen, Huu Hoang Nguyen, Long Le Thanh, Zahra Ahmadi, Thanh Nam Doan, Daoyuan Wu, Lingxiao Jiang
Research Collection School Of Computing and Information Systems
Detecting vulnerabilities in smart contracts is vital for the security and reliability of decentralized apps. To facilitate vulnerability detection, contract codes, including bug patterns, are represented as heterogeneous graphs with various nodes and edges, like control-flow and function-call graphs. However, existing graph learning techniques struggle with large, complex graphs. This paper presents MANDO-LLM, a novel framework that combines heterogeneous graph transformers (HGTs) with large language models (LLMs) for detecting vulnerabilities in smart contracts represented as heterogeneous contract graphs built upon control-flow and call graphs. MANDO-LLM uses LLMs to capture code features from control-flow and call data, customizes HGTs to learn …
Evaluating Defi Vulnerabilities: The Role Of Bug Bounty Programs On Defi Software Supply Chain, Ping Fan Ke, Yi Meng Lau, Lingxiao Jiang
Evaluating Defi Vulnerabilities: The Role Of Bug Bounty Programs On Defi Software Supply Chain, Ping Fan Ke, Yi Meng Lau, Lingxiao Jiang
Research Collection School Of Computing and Information Systems
Decentralized finance (DeFi), powered by blockchain technology, enables peer-to-peer financial transactions without intermediaries. Despite rapid adoption, DeFi attracts malicious actors exploiting vulnerabilities. To mitigate risks, we propose a framework assessing entry points in the DeFi software supply chain: smart contracts, oracles/third-party feeds, user interfaces, off-chain storage, and crypto wallets. Applying this framework, we evaluate whether industry solutions—particularly bug bounty programs—adequately address these gaps. Our preliminary analysis indicates that most programs cover smart contract vulnerabilities (85.7%), followed by user interface issues (21.3%) and crypto wallet loopholes (11.9%). However, third-party risks, such as oracle feeds, are frequently deemed out of scope. This …
A Socio-Technical Analysis Of Market Reactions On Meme Coins: Trump’S Presidential Effect, Ping Fan Ke, Yi Meng Lau
A Socio-Technical Analysis Of Market Reactions On Meme Coins: Trump’S Presidential Effect, Ping Fan Ke, Yi Meng Lau
Research Collection School Of Computing and Information Systems
Meme coins are a unique type of cryptocurrency whose value is shaped by internet culture and viral trends. This study introduces a socio-technical research model to examine key factors influencing meme coin dynamics and applies it to analyze market reactions to Donald Trump’s 2024 U.S. presidential election victory and inauguration, focusing on the $TRUMP meme coin and other politics-related meme coins, known as PolitiFi. Using a mixed-methods approach, we analyze publicly available news, social media activity, and marketplace data to investigate the interaction between social engagement and technical infrastructure. Econometric analysis shows that Trump-related events triggered short-term price surges, increased …
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
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 …
Misodice: Multi-Agent Imitation From Mixed-Quality Demonstrations, The Viet Bui, Tien Mai, Hong Thanh Nguyen
Misodice: Multi-Agent Imitation From Mixed-Quality Demonstrations, The Viet Bui, Tien Mai, Hong Thanh Nguyen
Research Collection School Of Computing and Information Systems
We study offline imitation learning (IL) in cooperative multi-agent settings, where demonstrations have unlabeled mixed quality - containing both expert and suboptimal trajectories. Our proposed solution is structured in two stages: trajectory labeling and multi-agent imitation learning, designed jointly to enable effective learning from heterogeneous, unlabeled data. In the first stage, we combine advances in large language models and preference-based reinforcement learning to construct a progressive labeling pipeline that distinguishes expert-quality trajectories. In the second stage, we introduce MisoDICE, a novel multi-agent IL algorithm that leverages these labels to learn robust policies while addressing the computational complexity of large joint …
No Experts, No Problem: Avoidance Learning From Bad Demonstrations, Minh Huy Hoang, Tien Mai, Pradeep Varakantham
No Experts, No Problem: Avoidance Learning From Bad Demonstrations, Minh Huy Hoang, Tien Mai, Pradeep Varakantham
Research Collection School Of Computing and Information Systems
This paper addresses the problem of learning avoidance behavior within the context of offline imitation learning. In contrast to conventional methodologies that prioritize the replication of expert or near-expert demonstrations, our work investigates a setting where expert (or desirable) data is absent, and the objective is to learn to eschew undesirable actions by leveraging demonstrations of such behavior (i.e., learning from negative examples).To address this challenge, we propose a novel training objective grounded in the maximum entropy principle. We further characterize the fundamental properties of this objective function, reformulating the learning process as a cooperative inverse Q-learning task. Moreover, we …
A Partition Cover Approach To Tokenization, Jia Peng Lim, Shawn Tan, Davin Choo, Hady Wirawan Lauw
A Partition Cover Approach To Tokenization, Jia Peng Lim, Shawn Tan, Davin Choo, Hady Wirawan Lauw
Research Collection School Of Computing and Information Systems
Tokenization is the process of encoding strings into tokens of a fixed vocabulary size, and is widely utilized in Natural Language Processing applications. The leading tokenization algorithm today is Byte Pair Encoding (BPE), which formulates the tokenization problem as a compression problem and tackles it by performing sequences of merges. In this work, we formulate tokenization as an optimization objective, show that it is NP-hard via a simple reduction from vertex cover, and propose a polynomial-time greedy algorithm GreedTok. Our formulation naturally relaxes to the well-studied weighted maximum coverage problem which has a simple -approximation algorithm GreedWMC. Through empirical evaluations …
When Less Language Is More: Language-Reasoning Disentanglement Makes Llms Better Multilingual Reasoners, Weixiang Zhao, Jiahe Guo, Yang Deng, Tongtong Wu, Wenxuan Zhang, Yulin Hu, Xingyu Sui, Yanyan Zhao, Wanxiang Che, Bing Qin, Tat-Seng Chua, Ting Liu
When Less Language Is More: Language-Reasoning Disentanglement Makes Llms Better Multilingual Reasoners, Weixiang Zhao, Jiahe Guo, Yang Deng, Tongtong Wu, Wenxuan Zhang, Yulin Hu, Xingyu Sui, Yanyan Zhao, Wanxiang Che, Bing Qin, Tat-Seng Chua, Ting Liu
Research Collection School Of Computing and Information Systems
Multilingual reasoning remains a significant challenge for large language models (LLMs), with performance disproportionately favoring high-resource languages. Drawing inspiration from cognitive neuroscience, which suggests that human reasoning functions largely independently of language processing, we hypothesize that LLMs similarly encode reasoning and language as separable components that can be disentangled to enhance multilingual reasoning. To evaluate this, we perform a causal intervention by ablating language-specific representations at inference time. Experiments on 10 open-weight LLMs spanning 11 typologically diverse languages show that this language-specific ablation consistently boosts multilingual reasoning performance. Layer-wise analyses further confirm that language and reasoning representations can be effectively …
Sheetpedia: A 300k-Spreadsheet Corpus For Spreadsheet Intelligence And Llm Fine-Tuning, Zailong Tian, Zhuoheng Han, Houfeng Wang, Lizi Liao
Sheetpedia: A 300k-Spreadsheet Corpus For Spreadsheet Intelligence And Llm Fine-Tuning, Zailong Tian, Zhuoheng Han, Houfeng Wang, Lizi Liao
Research Collection School Of Computing and Information Systems
Spreadsheets are widely used for data analysis and reporting, yet their complex structure and formula logic pose significant challenges for AI systems. We introduce Sheetpedia, a large-scale corpus of over 290,000 diverse spreadsheets (from 324,000+ workbooks) compiled from enterprise email archives and online forums. We detail a rigorous collection and preprocessing pipeline (integrating the Enron email spreadsheet archive and the Fuse web corpus, plus a new crawl of Excel forums) to standardize formats, filter languages, and remove duplicates. Sheetpedia provides extensive coverage of real formulas and annotations – addressing a gap left by prior table datasets (e.g. web tables used …
Temporal And Heterogeneous Graph Neural Network For Remaining Useful Life Prediction, Zhihao Wen, Yuan Fang, Pengcheng Wei, Fayao Liu, Zhenghua Chen, Min Wu
Temporal And Heterogeneous Graph Neural Network For Remaining Useful Life Prediction, Zhihao Wen, Yuan Fang, Pengcheng Wei, Fayao Liu, Zhenghua Chen, Min Wu
Research Collection School Of Computing and Information Systems
Predicting remaining useful life (RUL) plays a crucial role in the prognostics and health management of industrial systems that involve a variety of interrelated sensors. Given a constant stream of time-series sensory data from such systems, deep learning (DL) models have risen to prominence at identifying complex, nonlinear temporal dependencies in these data. In addition to the temporal dependencies of individual sensors, spatial dependencies emerge as important correlations among these sensors, which can be naturally modeled by a temporal graph that describes time-varying spatial relationships. However, the majority of existing studies have relied on capturing discrete snapshots of this temporal …