Computer Sciences Commons^™

Polymind: Parallel Visual Diagramming With Large Language Models To Support Prewriting Through Microtasks, Qian Wan, Jiannan Li, Huanchen Wang, Zhicong Lu

Research Collection School Of Computing and Information Systems

Prewriting is the process of generating and organising ideas before a first draft. It consists of a combination of informal, iterative, and semi-structured strategies such as visual diagramming, which poses a challenge for collaborating with large language models (LLMs) in a turn-taking conversational manner. We present Polymind, a visual diagramming tool that leverages multiple LLM-powered agents to support prewriting. The system features a parallel collaboration workflow in place of the turn-taking conversational interactions. It defines multiple ''microtasks'' to simulate group collaboration scenarios such as collaborative writing and group brainstorming. Instead of repetitively prompting a chatbot for various purposes, Polymind enables …

Emoshortcuts: Emotionally Expressive Body Augmentation For Social Mixed Reality Avatars, Hyuna Seo, Youngki Lee, Rajesh Krishna Balan, Thivya Kandappu

Research Collection School Of Computing and Information Systems

We present EmoShortcuts1, a novel social Mixed Reality (MR) framework that enhances emotional expression by dynamically augmenting avatar body gestures to reflect users’ emotional states. While social MR enables immersive remote interactions through avatars, conveying emotions remains challenging due to limitations in head-mounted display (HMD) tracking (e.g., missing lower-body movements, such as stomping or defensive postures), and users’ tendency to deprioritize nonverbal expressions during multitasking. EmoShortcuts addresses these challenges by introducing an augmentation framework that generates expressive body gestures even when users’ physical movements are restricted. We conducted a formative study with 12 participants to identify key challenges in emotional …

Boosting Chart-To-Code Generation In Mllm Via Dual Preference-Guided Refinement, Zhihan Zhang, Yixin Cao, Lizi Liao

Research Collection School Of Computing and Information Systems

Translating chart images into executable plotting scripts-referred to as the chart-to-code generation task-requires Multimodal Large Language Models (MLLMs) to perform fine-grained visual parsing, precise code synthesis, and robust cross-modal reasoning. However, this task is inherently under-constrained: multiple valid code implementations can produce the same visual chart, and evaluation must consider both code correctness and visual fidelity across diverse dimensions. This makes it difficult to learn accurate and generalizable mappings through standard supervised fine-tuning. To address these challenges, we propose a dual preference-guided refinement framework that combines a feedback-driven, dual-modality reward mechanism with iterative preference learning. Our approach introduces a structured …

Spd: Shallow Backdoor Protecting Deep Backdoor Against Backdoor Detection, Shunjie Yuan, Xinghua Li, Xuelin Cao, Haiyan Zhang, Mengyao Zhu, Robert H. Deng

Research Collection School Of Computing and Information Systems

Backdoor attacks have revealed the vulnerability of deep neural networks (DNNs), which motivates the development of secure deep learning systems. However, existing backdoor attacks often fail to bypass backdoor detection and human visual inspection, resulting in the exposure of the backdoor implanted in DNNs, which can subsequently be significantly mitigated through pruning or fine-tuning on benign data. To address this issue, in this paper, we propose a novel backdoor attack called SPD (Shallow Protecting Deep), which consists of a deep backdoor in the frequency domain and a shallow backdoor in the pixel domain, where the shallow backdoor acts as a …

Impact Of Original Versus Reposted Social Endorsements On Content Consumption: The Moderating Role Of Endorsers’ Network Characteristics, Anqi Zhao, Qian Tang

Research Collection School Of Computing and Information Systems

Social endorsements broadcast endorsers’ positive attitudes toward content or products, especially to their social ties. Original endorsements created by endorsers can be propagated further as reposted endorsements. Both are important marketing tools to increase content consumption, yet their differences are unclear. This study compares the impacts of original and reposted endorsements on content consumption and their contingencies on the endorsers’ network characteristics. Using data on social endorsements of YouTube videos on Twitter, we find that original endorsements (i.e., original tweets) significantly boost content consumption, and the effect is positively moderated by the endorsers’ network size but not their tie strength. …

Juxtaposing Approaches To Risk-Based Ai Governance In Different ‘Rights’ Contexts: A Comparative Analysis Between Singapore And The Eu, Jane Loo, Mark Findlay

Research Collection Yong Pung How School Of Law

Comparative analysis of European and certain Asian approaches to governance often degenerates into simplistic dichotomies based on universal human rights assumptions. This chapter rejects such dualities, ill-informed by theory and historical reflection. The emerging argument is founded on a historical realist approach to theorising difference. Assisted by Polanyi’s double movement, the detailed substantive comparison is preceded by considerations of how recent trends in governing AI have uniformly adopted a countermovement against the dis-embedding of data and technology from the social leading to a risk/responsibility paradigm. From here, a more nuanced reflection of AI governance approaches in the EU and Singapore …

Look Before You Decide: Prompting Active Deduction Of Mllms For Assumptive Reasoning, Yian Li, Wentao Tian, Yang Jiao, Jingjing Chen, Tianwen Qian, Bin Zhu, Na Zhao, Yu‑Gang Jiang

Research Collection School Of Computing and Information Systems

Recently, Multimodal Large Language Models (MLLMs) have achieved significant success across multiple disciplines due to their exceptional instruction-following capabilities and extensive world knowledge. However, whether these MLLMs possess human-like compositional reasoning abilities remains an open problem. To unveil their reasoning behaviors, we first curate a Multimodal Assumptive Reasoning Benchmark (MARS-Bench) in this paper. Interestingly, we find that most prevalent MLLMs can be easily fooled by the introduction of a presupposition into the question, whereas such presuppositions appear naive to human reasoning. Besides, we also propose a simple yet effective method, Active Deduction (AD), a novel reinforcement learning paradigm to encourage …

Memory-Efficient 4-Bit Preconditioned Stochastic Optimization, Jingyang Li, Kuangyu Ding, Kim-Chuan Toh, Pan Zhou

Research Collection School Of Computing and Information Systems

Preconditioned stochastic optimization algorithms, exemplified by Shampoo, outperform first-order optimizers by offering theoretical convergence benefits and practical gains in large-scale neural network training. However, they incur substantial memory overhead due to the storage demands of non-diagonal preconditioning matrices. To address this, we introduce 4-bit quantization for Shampoo’s preconditioners. We introduce two key methods: First, we apply Cholesky decomposition followed by quantization of the Cholesky factors, reducing memory usage by leveraging their lower triangular structure while better preserving spectral properties to minimize information loss. To our knowledge, this is the first quantization approach applied to Cholesky factors of preconditioners. Second, we …

What Students Really Think: Unpacking Ai Ethics In Educational Assessments Through A Triadic Framework, Lim Ming Soon Tristan, Gottipati Swapna, Michelle L. F. Cheong

Research Collection School Of Computing and Information Systems

The rise of AI in educational assessments has significantly enhanced efficiency and accuracy. However, it also introduces critical ethical challenges, including bias in grading, data privacy risks, and accountability gaps. These issues can undermine trust in AI-driven assessments and compromise educational fairness, making a structured ethical framework essential. To address these challenges, this study empirically validates an existing triadic ethical framework for AI-assisted educational assessments, originally proposed by Lim, Gottipati and Cheong (In: Keengwe (ed) Creative AI tools and ethical implications in teaching and learning, IGI Global, 2023), grounded in student perceptions. The framework encompasses three ethical domains—physical, cognitive, and …

Rethinking Teaching Evaluation Reports: Designing Ai-Transformed Student Feedback For Instructor Engagement, Ruoxi Shang, Keri Mallari, Au Wei Bin Yeong, Ken Yasuhara, Anthony Tang, Gary Hsieh

Research Collection School Of Computing and Information Systems

Student feedback is critical for improving teaching, yet instructors often avoid reading evaluations due to emotional burden and information overload. We present a systematic exploration of how language models can distill and transform student evaluations into adaptive, actionable insights. Through a systematic design space exploration combining 4 feedback strategies (removing harmful content, paraphrasing criticism, sandwiching negatives, adding constructive suggestions) with 4 presentation formats (themes, cards, letters, chatbots), we created six AI-augmented prototypes of teaching evaluations. Interviews with 16 post-secondary instructors revealed that effective use of AI in feedback processing should: (1) support action formation through focused views and divergent thinking, …

Conditional Attribute-Based Pre: Definition And Construction From Lwe, Lisha Yao, Jian Weng, Pengfei Wu, Guofeng Tang, Guomin Yang, Haiyang Xue, Robert H. Deng

Research Collection School Of Computing and Information Systems

Attribute-based proxy re-encryption (AB-PRE) is a crucial variant of proxy re-encryption. It allows a proxy with a re-encryption key to transform a delegator’s ciphertext associated with an access policy into another ciphertext associated with a new access policy, enabling delegatees with matching attributes to decrypt the transformed ciphertext. However, a key limitation of AB-PRE is that the delegator cannot control which ciphertexts are transformed. As a result, the proxy, once given the re-encryption key, indiscriminately transforms all ciphertexts, effectively switching their underlying policies—an issue known as the all-or-nothing problem. It limits the system’s flexibility and practicality in real-world use cases.In …

Fcad: Feature-Coupled Anisotropic Diffusion For Continuous Graph Learning, Amitoz Azad, Zhiyuan Zhang

Research Collection School Of Computing and Information Systems

In this work, we propose a novel continuous graph neural network called FCAD (Feature-Coupled Anisotropic Diffusion) for the task of node classification on graphs. Our approach is motivated by the success of feature-coupled anisotropic diffusion PDEs in multivalued image restoration. Our method introduces a total variation regularization-inspired anisotropic term to control diffusion between nodes and incorporates a learnable parameterization for feature coupling during the diffusion process. Our model performs competitively against several GNN baselines for both heterophilous and homophilous graphs, demonstrating notable benefits for heterophilous graphs due to the learnable feature coupling.

Lsfdnet: A Single-Stage Fusion And Detection Network For Ships Using Swir And Lwir, Yanyin Guo, Runxuan An, Junwei Li, Zhiyuan Zhang

Research Collection School Of Computing and Information Systems

Traditional ship detection methods primarily rely on single-modal approaches, such as visible or infrared images, which limit their application in complex scenarios involving varying lighting conditions and heavy fog. To address this issue, we explore the advantages of short-wave infrared (SWIR) and long-wave infrared (LWIR) in ship detection and propose a novel single-stage image fusion detection algorithm called LSFDNet. This algorithm leverages feature interaction between the image fusion and object detection subtask networks, achieving remarkable detection performance and generating visually impressive fused images. To further improve the saliency of objects in the fused images and improve the performance of the …

Tactile Data Comics: Combining Step-By-Step Presentation Of Tactile Graphics With Verbal Narration For The Blind And Visually Impaired, Yang Jiao, Ruoting Sun, Rong Luo, Xiwen Yao, Xinran She, Kotaro Hara, Yuewen Zhang, Xinyi Fu

Research Collection School Of Computing and Information Systems

Tactile graphics on a refreshable display have proven effective in enabling visually impaired people to comprehend pictorial content. To further evaluate the effectiveness of refreshable tactile displays in blind education, we designed tactile data comics, a method that combines step-by-step presentation of tactile graphics with verbal narration. We conducted a user study with sixteen visually impaired students to compare tactile data comics against verbal-only and static tactile graphics. Our findings show that tactile data comics significantly improve participants’ comprehension and engagement during the learning experience. These empirical results suggest that the integration of refreshable tactile displays and tactile data comics …

Language-Driven 3d Human Pose Estimation In Multi-Person Scenarios: A New Dataset And Approach, Tingrui Shen, Bangzhen Liu, Zhirun Fan, Shiting Zhang, Weifeng Pan, Sun Fan, Dan Cao, Shengfeng He

Research Collection School Of Computing and Information Systems

In an NBA game scenario, consider the challenge of locating and analyzing the 3D poses of players performing a user-specified action, such as attempting a shot. Traditional 3D human pose estimation (3DHPE) methods often fall short in such complex, multi-person scenes due to their lack of semantic integration and reliance on isolated pose data. To address these limitations, we introduce Language-Driven 3D Human Pose Estimation (L3DHPE), a novel approach that extends 3DHPE to general multi-person contexts by incorporating detailed language descriptions. We present Panoptic-L3D, the first dataset designed for L3DHPE, featuring 3,838 linguistic annotations for 1,476 individuals across 588 videos, …

Information-Bottleneck Driven Binary Neural Network For Change Detection, Kaijie Yin, Zhiyuan Zhang, Shu Kong, Tian Gao, Cheng-Zhong Xu, Hui Kong

Research Collection School Of Computing and Information Systems

In this paper, we propose Binarized Change Detection (BiCD), the first binary neural network (BNN) designed specifically for change detection. Conventional network binarization approaches, which directly quantize both weights and activations in change detection models, severely limit the network's ability to represent input data and distinguish between changed and unchanged regions. This results in significantly lower detection accuracy compared to real-valued networks. To overcome these challenges, BiCD enhances both the representational power and feature separability of BNNs, improving detection performance. Specifically, we introduce an auxiliary objective based on the Information Bottleneck (IB) principle, guiding the encoder to retain essential input …

Developing A Strong Cps Defender: An Evolutionary Approach, Qingyuan Hu, Christopher M. Poskitt, Jun Sun, Yuqi Chen

Research Collection School Of Computing and Information Systems

Cyber-physical systems (CPSs) are used extensively in critical infrastructure, underscoring the need for anomaly detection systems that are able to catch even the most motivated attackers. Traditional anomaly detection techniques typically do `one-off' training on datasets crafted by experts or generated by fuzzers, potentially limiting their ability to generalize to unseen and more subtle attack strategies. Stopping at this point misses a key opportunity: a defender can actively challenge the attacker to find more nuanced attacks, which in turn can lead to more effective detection capabilities. Building on this concept, we propose Evo-Defender, an evolutionary framework that iteratively strengthens CPS …

Hlcg: A Hierarchical Lane-Changing Gaming Decision Model For Heterogeneous Traffic Flow On Two-Lane Highways, Tianyi Wang, Chong He, Hao Li, Yixuan Li, Yiming Xu, Yangyang Wang, Junfeng Jiao

Research Collection College of Integrative Studies

Discretionary lane-changing behavior is one of the most common highway operations, which seriously affects traffic efficiency and safety. Nowadays, connected and automated vehicles (CAVs) are advancing rapidly, though not yet fully widespread. As a result, a mixed traffic environment with traditional human-driven vehicles (HDVs) and CAVs will persist for the foreseeable future. To achieve effective automatic lane-changing maneuvers, it’s necessary to propose a lane-changing decision model for heterogeneous traffic flow on two-lane highways. This paper firstly extends longitudinal car-following models based on the intelligent driver model and lateral lane-changing models using quintic polynomial curves to accommodate heterogeneous traffic flow, and …

Towards A Digital Twin For Smart Resilient Cities: Real-Time Fire And Smoke Tracking And Prediction Platform For Community Awareness (Firecom), Kijin Seong, Junfeng Jiao, Ryan Lewis Hardesty, Arya Farahi, Paul Navratil, Nate Casebeer, Braniff Davis, Justice Jones, Dev Niyogi

Research Collection College of Integrative Studies

This paper discusses the development and application of a digital twin (DT) for urban resilience, focusing on an integrated platform for real-time fire and smoke. The proposed platform, FireCom, adapts DT concepts for the unique challenges of urban fire management, which differ significantly from regional wildfire systems. Through an exploratory case study in Austin, Texas, in the United States, this research bridges the theoretical foundations of 3D DT with their practical application in fire and smoke management. By fusing diverse data sources, ranging from air quality sensors and meteorological data to 3D urban infrastructure, FireCom supports both emergency response and …

Application Of Machine Learning For Vascular System Analysis, Alireza Bagheri Rajeoni

Theses and Dissertations

The analysis of vascular structures is critical for diagnosing, monitoring, and treating vascular diseases such as aneurysms, stenosis, and vascular calcification. Traditional methods often rely on manual interpretation of imaging data, which is time-consuming, subjective, and not scalable. This work explores the application of advanced machine learning techniques to automate and enhance vascular system analysis. Our contributions include achieving state-of-the-art accuracy in vascular segmentation, developing a machine learning pipeline to automatically quantify vascular calcification in peripheral arterial disease, and designing a multi-stage machine learning system for abdominal aortic aneurysm analysis that identifies aneurysm boundaries and estimates aneurysm volume in a …

Implicit Neural Representation For Image Reconstruction, Canyu Zhang

Theses and Dissertations

Image reconstruction seeks to restore corrupted images and recover visual content that has been lost or degraded. Such degradation may result from low resolution, occlusion, masking, or shadow interference. This problem has become an increasingly significant research topic, as visual information plays a central role in almost every aspect of modern life. Neural network based approaches have recently emerged as highly effective solutions for this task. In particular, convolutional neural networks and transformer based architectures have demonstrated remarkable success in producing visually convincing reconstructions. However, these models remain constrained in several important ways, one of the most critical being that …

Governance In The Absence Of Government, Tracy Hresko Pearl

Faculty Articles

Artificial intelligence (AI) is advancing at an unprecedented pace, with generative systems exerting growing influence over social, economic, and political life. While Al offers opportunities for innovation and efficiency, it also poses risks ranging from misinformation and job displacement to existential threats if highly autonomous systems evade human control. Across industry, government, and civil society, there is broad consensus that Al requires oversight.

Yet traditional U.S. regulatory approaches face six significant barriers: (1) technology outpacing legislation, (2) limited Al expertise among policymakers, (3) regulatory capture, (4) political gridlock, (5) outdated governance structures, and (6) the inherent complexity of Al. Combined …

A System Framework To Symbolically Explore Intel Tdx Module Execution, Pansilu Pitigalaarachchillage, Xuhua Ding

Research Collection School Of Computing and Information Systems

We present TDXplorer, the first dynamic symbolic analysis system for Intel's TDX Module, the software trusted computing base of TDX. Without using TDX hardware, an analyzer function on top of TDXplorer can not only apply dynamic analysis to control and instrument the TDX Module's execution, but also carry out symbolic execution for path exploration as well as security and functionality reasoning. The two types of analysis are seamlessly integrated in a way that symbolic execution is conducted directly upon the TDX Module's binary code and runtime states, which are shaped by using dynamic analysis techniques. We implement TDXplorer on Linux …

Parameter-Efficient Variational Autoencoder For Multimodal Multi-Interest Recommendation, Nhu Thuat Tran, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

Learning user preferences in recommendation systems is enriched by multimodal features, such as textual and visual content, and amplified by multi-interest modeling with Variational AutoEncoders (VAEs). However, prior efforts are limited by single modality focus and cumbersome, parameter-heavy architecture designs. To address these limitations, we introduce an innovative solution that blends the semantic richness of multimodal data with the representational power of multi-representation VAEs. Drawing inspiration from Mixture of Experts (MoE), we cast each VAE as an expert tailored to a specific modality, then fuse them via a novel parameter-merging function into a lean, unified model. This approach efficiently captures …

Ivycross: A Privacy-Preserving And Concurrency Control Framework For Blockchain Interoperability, Ming Li, Jian Weng, Jia-Si Weng, Yi Li, Yongdong Wu, Dingcheng Li, Guowen Xu, Deng, Robert H.

Research Collection School Of Computing and Information Systems

Interoperability is a fundamental challenge for long-envisioned blockchain applications. A mainstream approach is using Trusted Execution Environment (TEE) to support interoperable off-chain execution. However, this incurs multiple TEE configured with non-trivial storage capabilities running on fragile concurrent processing environments, rendering current strategies based on TEE far from being practical. This paper aims to fill this gap and design a practical interoperability mechanism with simplified TEE as the underlying architecture. Specifically, we present IvyCross, a TEE-based framework that achieves low-cost, privacy-preserving, and race-free blockchain interoperability. IvyCross allows running arbitrary smart contracts across heterogeneous blockchains atop two distributed TEE-powered hosts. We design …

Morphology-Aware Hrv Estimation From Wrist Ppg In Sedentary Scenarios, Changshuo Hu, Hung Manh Pham, Dong Ma

Research Collection School Of Computing and Information Systems

Photoplethysmography (PPG) is widely used in wearable devices for non-invasive heart rate variability (HRV) monitoring. While most prior work focuses on mitigating motion artifacts, recent studies highlight that even subtle contact pressure variations can distort waveform morphology and lead to inaccurate HRV estimates. In this work, we propose a morphology-aware deep learning framework that conditions HRV estimation on beat-level waveform types. Our model jointly encodes the raw PPG waveform and a sequence of pressure-induced morphology labels using parallel encoders, integrates them via cross-attention, and predicts normal-to-normal (NN) intervals and beat count to support downstream HRV computation. Evaluated on the public …

Multi-Period Risk-Aware Procurement Optimization Under Covid-19 Disruption, Jonathan Chase, Hoong Chuin Lau, Jinfeng Yang, Lu Liu

Research Collection School Of Computing and Information Systems

Supply chain resilience has been a topic of active research in the operations research and AI communities for several years, but the COVID-19 pandemic threw the frailties of global supply chains into sharp relief. Disruptions and delays caused by fresh outbreaks leading to lockdowns, put severe strain on supply chains in many industries. In this work we develop lockdown-resilient procurement capabilities for a global technology company. First, through analysis of lockdown data from China we develop a logarithmic regression-based lockdown prediction method to complement a supplier risk metric for conventional risks. Second, we develop a multi-period stochastic optimization model that …

Lightweight Population-Based Policy Optimization For Pickup And Delivery Problems, Yizhou Liu, Li Li, Yixin Xu, Tang Liu, Rong Cheng, Die Wu, Jilin Yang, Jingwen Li

Research Collection School Of Computing and Information Systems

In recent years, applying deep models to automatically learn construction heuristics for vehicle routing problems has achieved remarkable advancements. However, they are less effective in searching solutions due to two primary limitations: relying on deterministic probability distributions and overlooking the strategic advantage of prioritizing nearby unvisited nodes during the route construction process, resulting in suboptimal policies In this paper, we propose a novel lightweight population-based policy optimization (LPPO) framework that learns a diverse population of solution strategies through the utilization of innovative perturbation factors, in order to facilitate search exploration. Moreover, we design a localized attention synthesis (LAS) network to …

From Holistic To Localized: Local Enhanced Adapters For Efficient Visual Instruction Fine-Tuning, Pengkun Jiao, Bin Zhu, Jingjing Chen, Chong-Wah Ngo, Yugang Jiang

Research Collection School Of Computing and Information Systems

Efficient Visual Instruction Fine-Tuning (EVIT) seeks to adapt Multimodal Large Language Models (MLLMs) to downstream tasks with minimal computational overhead. However, as task diversity and complexity increase, EVIT faces significant challenges in resolving data conflicts. To address this limitation, we propose the Dual Low-Rank Adaptation (Dual-LoRA), a holistic-to-local framework that enhances the adapter’s capacity to address data conflict through dual structural optimization. Specifically, we utilize two subspaces: a skill space for stable, holistic knowledge retention, and a rank-rectified task space that locally activates the holistic knowledge. Additionally, we introduce Visual Cue Enhancement (VCE), a multi-level local feature aggregation module designed …

Exploring Object Status Recognition For Recipe Progress Tracking In Non-Visual Cooking, Franklin Mingzhe Li, Kaitlyn Ng, Bin Zhu, Patrick Carrington

Research Collection School Of Computing and Information Systems

Cooking plays a vital role in everyday independence and well-being, yet remains challenging for people with vision impairments due to limited support for tracking progress and receiving contextual feedback. Object status — the condition or transformation of ingredients and tools — offers a promising but underexplored foundation for context-aware cooking support. In this paper, we present OSCAR (Object Status Context Awareness for Recipes), a technical pipeline that explores the use of object status recognition to enable recipe progress tracking in non-visual cooking. OSCAR integrates recipe parsing, object status extraction, visual alignment with cooking steps, and time-causal modeling to support real-time …