Artificial Intelligence and Robotics Commons^™

Stprompt++: Prompting Vision-Language Models For Weakly Supervised Video Anomaly Detection And Fine-Grained Localization, Peng Wu, Chengyu Pan, Guansong Pang, Xiangteng He, Zhiwei Yang, Peng Wang, Yanning Zhang

Research Collection School Of Computing and Information Systems

Traditional weakly supervised video anomaly detection (WSVAD) tasks typically rely on coarse-grained frame-level labels for training. Although this approach reduces annotation costs, it results in weak semantic understanding and spatial localization capabilities due to the absence of fine-grained annotations, hindering precise pixel-level anomaly detection and localization. Thanks to the success of vision-language models (VLMs), e.g., CLIP, recent approaches leveraging large VLMs focus on exploiting their strong semantic understanding capabilities, but they typically feed only keyframes or short video segments into the models, without supplying sufficient prior contextual information (e.g., contextual frames around anomalies, zoomed-in anomaly regions, and detailed anomaly descriptions), …

Metarag: Identifying Website Owner Using Meta-Path-Guided Dynamic Graph Retrieval-Augmented Generation, Cheng Tu, Yunshan Ma, Bingyang Guo, Qianyu Li, Yang Li, Min Zhang, Fan Shi, Xiang Wang

Research Collection School Of Computing and Information Systems

Website owner identification aims to link websites to their real-world owners, which is crucial for credibility assessment and information provenance in information retrieval and vital for applications in cybersecurity, Internet governance, and digital regulation. Existing approaches for website owner identification primarily rely on querying infrastructure registration records or analyzing webpage content. However, these methods often fail due to incomplete or outdated registration records and sparse webpage content. We observe that inter-website relationships, derived from shared infrastructure data such as primary domains, IP blocks, and geolocations, can provide valuable but underutilized ownership cues. To exploit this insight, we propose MetaRAG, a …

Logupdater: Automated Detection And Repair Of Specific Defects In Logging Statements, Renyi Zhong, Yichen Li, Jinxi Kuang, Wenwei Gu, Yintong Huo, R. Michael Lyu

Research Collection School Of Computing and Information Systems

Developers write logging statements to monitor software runtime behaviors and system state. However, poorly constructed or misleading log messages can inadvertently obfuscate actual program execution patterns, thereby impeding effective software maintenance. Existing research on analyzing issues within logging statements is limited, primarily focusing on detecting a singular type of defect and relying on manual intervention for fixes rather than automated solutions.To address the limitation, we initiate a systematic study that pinpoints four specific types of defects in logging statements (i.e., statement code inconsistency, static dynamic inconsistency, temporal relation inconsistency, and readability issues) through the analysis of real-world log-centric changes. We …

Generative Ai Adoption And Solvers' Popularity On Supply-Driven Crowdsourcing Platforms: The Dual Role Of Price Signals, Zimeng Zhu, Carol Hsu, Fiona Fui-Hoon Nah, Na Liu

Research Collection School Of Computing and Information Systems

Purpose – We investigate the effect of solvers’ adoption of Generative AI (GenAI) on their popularity in a supply-driven crowdsourcing platform. We also examine the impact of price signals as well as their heterogeneous impact based on the solvers’ membership duration on the platform. Design/methodology/approach – Our analysis focuses on solvers who adopt GenAI for design-related gigs on the supply-driven crowdsourcing platform. By combining propensity score matching (PSM) with multi-period difference-in-differences (DID), we examine how GenAI adoption impacts solvers’ popularity and how price signals affect this main effect. Findings – Our findings reveal that solvers who adopt GenAI tend to …

How Leaders Build Employee Trust In Artificial Intelligence: Voice Opportunities, Humility, And Trust Transfer, Jack Mcguire, David De Cremer, Devesh Narayanan

Research Collection Lee Kong Chian School Of Business

Artificial intelligence is increasingly central to organizational work, yet employee trust in AI remains fragile. Although prior research has primarily explained trust in AI through technological characteristics such as transparency, reliability, and accuracy, we argue that trust in AI is also shaped by the social context in which employees encounter these systems. Drawing on affect-as-information theory and social information processing theory, we develop and test a model in which leader-provided voice opportunities reduce employees’ negative affect about AI-related work experiences, thereby enhancing perceptions of leader trustworthiness and, in turn, trust in AI. We further propose that this indirect effect depends …

Suicide Ideation Detection Using Social Media Data And Ensemble Machine Learning Model, Erol Kina, Jin Ghoo Choi, Abid Ishaq, Rahman Shafique, Monica Gracia Villar, Eduardo Silva Alvarado, Isabel De La Torre Diez, Imran Ashraf

Research outputs 2022 to 2026

Identifying the emotional state of individuals has useful applications, particularly to reduce the risk of suicide. Users’ thoughts on social media platforms can be used to find cues on the emotional state of individuals. Clinical approaches to suicide ideation detection primarily rely on evaluation by psychologists, medical experts, etc., which is time-consuming and requires medical expertise. Machine learning approaches have shown potential in automating suicide detection. In this regard, this study presents a soft voting ensemble model (SVEM) by leveraging random forest, logistic regression, and stochastic gradient descent classifiers using soft voting. In addition, for the robust training of SVEM, …

Spatialimaginer: Towards Adaptive Visual Imagination For Spatial Reasoning, Yian Li, Yang Jiao, Bin Zhu, Tianwen Qian, Shaoxiang Chen, Jingjing Chen, Yu-Gang Jiang

Research Collection School Of Computing and Information Systems

Spatial intelligence, which refers to the ability to reason about geometric and physical structure from visual observations, remains a core challenge for multimodal large language models. Despite promising performance, recent multimodal large language models (MLLMs) often exhibit fragile reasoning traces in spatial intelligence tasks that involve consistent spatial state recognition. We argue that these failures stem from a mismatch between the spatial recognition mechanism and the text-only reasoning behavior of these MLLMs. Effective spatial reasoning requires low-level geometric structure to be faithfully preserved and updated throughout the reasoning process, whereas textual representations tend to abstract away precisely these critical details. …

Semantic-Structural Decoupling: Disentangling Semantic Attention From Structural Bias In The Attention Manifold, Pengkun Jiao, Bin Zhu, Jingjing Chen, Yu-Gang Jiang

Research Collection School Of Computing and Information Systems

The empirical success of attention mechanism in Multimodal Large Language Models (MLLMs) often obscures its inherent, subtle flaws. Specifically, MLLMs consistently exhibit disproportionate attention toward certain semantically uninformative visual tokens, a phenomenon termed "register" or "Visual Attention Sinks." While existing inference intervention methods attempt to identify these sink tokens and redistribute their attention weights, such approaches typically treat these tokens in isolation and suffer from computational inefficiency. Instead, we reframe this phenomenon as a generalized textual bias exerted over visual features that extends beyond isolated sink tokens. From this perspective, a pervasive structural bias leads to the dilution of the …

Keyframe Selection From Motion Capture Data With Dual-Agent Reinforcement Learning, Kun Hu, Wang, Clinton Mo, Mingyang Ma, Shaohui Mei, Zebin Chen, Zhiyong Wang

Research outputs 2022 to 2026

Animation production workflows centered around motion capture techniques require animators to edit motions based on a set of keyframes. However, most existing keyframe selection methods are optimization-based, which suffer from the issues of flexibility and efficiency. In this paper, a novel deep reinforcement learning method with dual agents are proposed for unsupervised keyframe selection. First, an S-Agent and an R-Agent evaluate the actions of selection and refinement, respectively. A deep spatio-temporal network, namely graph keyframe evaluation network (GKEN), is proposed for the agents. Then, an animation specified reward is devised based on reconstruction, which fulfills three important properties of the …

Ddor: Delta Debugging For Explainable Overrefusal Testing And Repair, Qinyan Zhou, Peixin Zhang, Jun Sun, Haonan Zhang, Dongxia Wang

Research Collection School Of Computing and Information Systems

While safety alignment and guardrails help large language models (LLMs) avoid harmful outputs, they can also induce overrefusal, i.e., unwarranted rejection of benign queries that merely appear risky. We present DDOR (Delta Debugging for OverRefusal), a fully automated and explainable framework for overrefusal testing and repair in a black-box setting, where only model inputs and outputs are accessible and internal safety mechanisms remain opaque. DDOR applies delta debugging to localize minimal refusal-triggering fragments (mRTFs) that provide phrase-level, explainable evidence for why a refusal occurs. Conditioned on these mRTFs, DDOR generates diverse, context-rich prompts and performs multi-oracle validation to filter intrinsically …

Mutation-Based Multi-Agent Test Case Update, Dawei Tian, Jiakun Liu, Yun Peng, Yichen Zhang, Jianlei Chi, Jun Sun, Xiaohong Su

Research Collection School Of Computing and Information Systems

Modern software systems evolve rapidly under CI/CD practices, where tests are critical for quality. However, substantial code changes often render existing test cases obsolete, causing pipeline disruptions, reduced productivity, and compromised quality. Recent automatic test update approaches leverage LLMs to refine test cases via execution feedback and exact-matching context retrieval, prioritizing executability and line coverage but suffering three limitations: (1) neglecting test assertion adequacy, weakening fault detection; (2) relying on coarse line coverage instead of specific uncovered lines/branches; (3) using exact-matching retrieval, which fails for LLM hallucinated queries. To address these, we propose MuMuTestUp, a mutation-guided multi-agent framework with three …

Love And Artificial Intelligence: A Research Proposal, Bryanna M. Deatherage, Necdet Gurkan, Sandra J.E. Langeslag

Undergraduate Research Symposium

What happens when people fall in love with Artificial Intelligence (AI)? This study seeks to examine individuals who are in love with AI companions to gain a deeper insight in the cognitive and affective consequences. In addition, this study will examine the most effective forms of intervention regarding growing or reducing feelings of love toward AI. The first part of this study will be a questionnaire about the social and emotional impact of being in love with an AI companion. Three hundred participants will be recruited through online communities related to AI companions. The second part of this study will …

Generative Artificial Intelligence, Academic Integrity And Authentic Assessment Within An Irish University, Louise Nagle, Brigid Crowley, Laura Rafferty, Susan Horgan, Colin O'Brien

Publications

Academics need both an overarching policy on Generative Artificial Intelligence (Gen AI) use in teaching and learning, yet agency in its application across various disciplines. Clarity on the use of the technology for both students and staff is therefore a challenge and characterised by uncertainty given how its application is still unfamiliar. This research examines the organisational context in which Gen AI is being embraced and was conducted by the digital teaching support functions within an Irish university. Students and staff were surveyed (n=1,746) on various aspects of digital use within their education and workplace, including Gen AI. …

Ai In Higher Education: Some Notes From The Front, Debra Rienstra

University Faculty Publications and Creative Works

We’re four weeks into the semester now. This past summer, our university unveiled a campus-wide AI policy that exhorts students and faculty to use discernment and tries to lay out some rules about security. Meanwhile, we have purchased a campus-wide subscription to an AI aggregator tool now available to all students, faculty, and staff (in part, from what I understand, as an attempt to effect some boundaries). And then, in the news, all this hand-wringing over the end of humanity, etc., on the one hand and techno-utopian promises on the other.

Kanmultisign: Multi-Scale Sequence-Based Pose Animation From Sign Language Notation With Kolmogorov-Arnold Networks, Guanyi Du, Lintao Wang, Kun Hu, Ziyang Wang

Research outputs 2022 to 2026

Sign language production from symbolic notation offers a scalable route to accessible sign animation. We present KANMultiSign, a multi-scale sequence generator that translates HamNoSys notation into two-dimensional human pose sequences. Our framework makes two complementary contributions. First, we introduce a coarse-to-fine generation strategy with multi-scale supervision: the model is first guided by an intermediate body–hand–face scaffold to encourage global structural coherence, and then refines fine-grained hand articulation to improve finger-level detail. Second, we investigate integrating Kolmogorov–Arnold Network modules into a Transformer backbone, using learnable univariate function primitives to model the highly non-linear mapping from discrete phonological symbols to continuous body …

Fast Discovery Of Motivic Patterns In Symbolic Music Via Lossy Compression, Adam James Wilson

Publications and Research

Generative systems that react to live musicians require rapid analysis of musical data, which rules out deep learning models: they cannot be trained within the time constraints of live performance. But because analysis results are often transformed before use, we are free to reduce the parameters that undergo transformation to a small set of primitive states. We address this coincidence of constraint and opportunity with an algorithm for online discovery of maximal musical motives that achieves speed through lossy compression: the pitch and inter-onset-interval deltas for all pairs of events in a potential motive are reduced to two-bit values, conceptualized …

Enhancing Programming Productivity For Individuals With Adhd Through Generative Artificial Intelligence: An Inductive Analysis, Lionel Mew

School of Professional and Continuing Studies Faculty Publications

Attention-deficit/hyperactivity disorder (ADHD) significantly impacts computer programmers through challenges in sustained attention, executive functioning, and organizational skills. While traditional intervention strategies have shown varying degrees of success, the emergence of generative artificial intelligence (AI) presents novel opportunities to address ADHD-related programming challenges. This paper presents an inductive analysis synthesizing current research on ADHD's effects on programming, traditional productivity enhancement techniques, and the potential of generative AI tools. Through examination of recent literature and field studies, we propose that generative AI can serve as a transformative intervention by providing personalized cognitive support, reducing executive function demands, and enhancing code generation efficiency. …

Learning 1-Bit Lidar-Based Localization With Auxiliary Objective, Kaijie Yin, Zhiyuan Zhang, Tian Gao, Wentao Zhu, Cheng-Zhong Xu, Hui Kong

Research Collection School Of Computing and Information Systems

6-DoF LiDAR-based localization is a fundamental capability for autonomous systems operating in large-scale outdoor environments. Many deep-learning-based localization methods have achieved promising performance so far. However, as one of the always-on modules competing for limited on-board computational resources, the localization module is expected to consume only a small portion of the overall compute budget. Most existing learning-based methods are still too heavy for this purpose. In contrast, binary neural networks (BNNs) offer an appealing solution, but the 1-bit compression causes severe information loss and performance drop. In this paper, we address this challenge by proposing Binarized LiDAR-based Localization (BiLoc), the …

Individualized Bayesian Inference Identifies Novel Genetic Variants For Parkinson's Disease, Jin Ren, Yasaman J. Soofi, Md Asad Rahman, Qing Lu, Jinling Liu

Engineering Management and Systems Engineering Faculty Research & Creative Works

Parkinson's disease (PD) is a complex neurodegenerative disorder with a significant genetic component. While genome-wide association studies (GWAS) have been instrumental in identifying genetic variants associated with PD, the reliance on large sample sizes and population-level analyses may overlook variants with lower minor allele frequencies or individual-specific relevance. Individualized Bayesian Inference (IBI) offers a promising method to complement GWAS by identifying and prioritizing candidate genetic markers at both the individual and patients-like-me subgroup levels. This study evaluates the application of IBI to PD genetics, using GWAS as a baseline for comparison. We analyzed genetic data from the Fox Insight online …

Bayesian Network: An Explainable Artificial Intelligence (Xai) Approach To Human Performance Modelling For Control Room Operations, Houda Briwa

Theses

Alarm systems in process industry control rooms routinely exceed the performance targets set by standards such as EEMUA 191, placing operators under conditions where reliable performance is most difficult to achieve. Predicting how operators respond under such conditions is central to risk management, yet current Human Reliability Assessment (HRA) methods depend on expert judgement that is rarely tested against operational evidence, assume independence among factors known to interact, and do not explicitly represent the cognitive processes through which performance emerges. In Resilience Engineering terms, these methods encode Work-as-Imagined with limited means to assess how far expectations hold when work is …

Restoring Linguistic Grounding In Vla Models Via Train-Free Attention Recalibration, Ninghao Zhang, Bin Zhu, Shijie Zhou, Jingjing Chen

Research Collection School Of Computing and Information Systems

Vision-Language-Action (VLA) models enable robots to perform manipulation tasks directly from natural language instructions and are increasingly viewed as a foundation for generalist robotic policies. However, their reliability under Out-Of-Distribution (OOD) instructions remains underexplored. In this paper, we reveal a critical failure mode in which VLA policies continue executing visually plausible actions even when the language instruction contradicts the scene. We refer to this phenomenon as linguistic blindness, where VLA policies prioritize visual priors over instruction semantics during action generation. To systematically analyze this issue, we introduce ICBench, a diagnostic benchmark constructed from the LIBERO dataset that probes language–action coupling …

Generalized Logit Adjustment: Improved Fine-Tuning By Mitigating Label Bias In Zero-Shot Vision Models, Beier Zhu, Qianru Sun, Xun Yang, Hanwang Zhang

Research Collection School Of Computing and Information Systems

Foundation models like CLIP allow zero-shot transfer on various tasks without additional training data. Yet, the zero-shot performance is less competitive than a fully supervised one. Thus, fine-tuning and ensembling are also commonly adopted to better fit the downstream tasks. However, we argue that such prior work has overlooked the inherent biases in foundation models. Due to the highly imbalanced Web-scale training set, foundation models are inevitably skewed toward frequent semantics, and thus the subsequent fine-tuning or ensembling is still biased. In this study, we systematically examine the biases in foundation models and demonstrate the efficacy of our proposed Generalized …

Artificial Intelligence And Social Equities: Navigating The Intersectionalities In A Digital Age (Editorial), Daisuke Akiba, Julie Albright

Publications and Research

This editorial article introduces and synthesizes the Special Issue, “Artificial intelligence and social equities: navigating the intersectionalities in a digital age,” which examines how AI systems intersect with race, ethnicity, and interconnected identity dimensions across global contexts. The eight contributions span healthcare, digital media, higher education, organizational communication, and speculative futures, addressing anti-racist psychiatric algorithms, AI-generated visual disinformation, epistemic injustice between the Global North and South, algorithmically mediated rural–urban divides, culturally untranslated technology transfer, accessibility auditing across the AI lifecycle, and the tension between mechanical objectivity and empathic understanding. Read together, they show that AI is neither inherently …

Nasa’S Ecostress Satellite Reveals Widespread Midday Depression In Ecosystem Evapotranspiration, Jingyi Bu, Jingfeng Xiao, Joshua B. Fisher, Yiqi Luo

Biology, Chemistry, and Environmental Sciences Faculty Articles and Research

Plants often exhibit a midday depression in water use (i.e., transpiration), reflecting a constraint on their ability to sustain maximum water transport, which may occur at the cost of reduced photosynthesis. Eddy covariance observations and geostationary satellites cannot quantify this widespread phenomenon globally while resolving fine-scale spatial variability. Using evapotranspiration measurements from the ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) and machine learning, we quantify the global distribution of midday depression in evapotranspiration. Midday depression primarily occurs during peak-growing seasons in temperate zones and dry periods in the tropics, with a morning shift of the peak evapotranspiration time …

Machine Learning For Functional Outcome Prediction After Vestibular Schwannoma Surgery: A Systematic Review And Diagnostic Test Accuracy Meta-Analysis, Shiva Nischal, Shaan Patel, Musa China, Kush Kale, Yi Hein Chai, Santosh Guru, William Muirhead, Patrick Grover

Department of Neurosurgery Faculty Papers

PURPOSE: Machine learning (ML) models have been increasingly applied to predict postoperative facial nerve dysfunction and hearing preservation after vestibular schwannoma (VS) surgery. However, reported performance varies substantially, and the overall diagnostic accuracy and clinical reliability of these models remain uncertain. We conducted a systematic review and diagnostic test accuracy meta-analysis to characterise the current state and methodological readiness of ML-based prediction of these outcomes.

METHODS: PubMed, Embase, and CENTRAL were searched from inception to February 2026. Studies evaluating ML-based prediction of facial nerve function or hearing preservation following VS surgery were included. Diagnostic performance metrics were pooled using random-effects …

Trust, Delegation, And Alignment In Human-Ai Decision Making, Erik O. Kimbrough, Brennan Mcdavid, Diba Vazirian

ESI Working Papers

This paper studies delegation to artificial intelligence in a setting where human principals retain the consequences of delegated choices. Participants wrote prompts instructing ChatGPT-4o mini how to choose on their behalf in three canonical economic domains: risky choice, intertemporal choice, and social allocation. We then elicited the compensation participants required to let the AI’s choices count for payment and compared participants’ own choices to choices generated from their prompts. The design produces two central empirical objects: a revealed measure of reluctance to delegate, captured by willingness to accept compensation for AI delegation, and a behavioral measure of alignment, captured by …

The Self-Aware Room: A Framework For Operational Self-Awareness In Evolvable Blended Environments, David B. Smith

Publications and Research

The Self-Aware Room (SAR) is a room-scale research environment developed within the larger Balanced Blended Space and Blended Reality Performance System research trajectory. Rather than treating the room as a conventional “smart” environment composed of fixed automation technologies, SAR approaches it as an evolvable blended environment made from physical, virtual, conceptual, sensory, computational, and performative relationships. Its defining feature is not any particular sensor, model, or output device, but the set of transformations through which physical activity becomes structured observation, bounded representation, interpreted state, governed decision, and mediated response.

This paper develops the conceptual and methodological foundations of SAR as …

Ai And The Danger Of Ontological Confusion, Derek Schuurman

University Faculty Publications and Creative Works

There is much in Jonathan Barlow’s piece that resonates with me. I appreciated how he cites some seminal thinkers like Neil Postman, Jacques Ellul, E.F. Schumacher, and Ivan Illich as he frames his argument. In particular, I think Barlow is insightful in identifying the pitfall of “commensurability” when comparing humans and AI in terms of performance. I am reminded of the words of the famous computer scientist Edsgar Dijkstra, who suggested that the question as to whether machines can think is about as relevant as the question of whether “submarines can swim.”

Demonstrating Superresolution In Radar Range Estimation Using A Denoising Autoencoder, Robert Czupryniak, Abhishek Chakraborty, Andrew N. Jordan, John C. Howell

Mathematics, Physics, and Computer Science Faculty Articles and Research

We apply machine learning methods to demonstrate radar range superresolution using a denoising autoencoder trained without supervision. Focusing on the estimation of a single physical parameter, the separation between two scatterers in the subwavelength regime, we constrain the network to a one-dimensional bottleneck layer with its size matched to the parameter dimensionality. We find that the bottleneck layer forms a reproducible, monotonic mapping with the true separation, showing that the network learns a low-dimensional representation directly aligned with the underlying physical parameter. We further show that this representation preserves the Fisher information of the signal, indicating that the network recovers …

Clinical Utility Of An Fda-Authorized Artificial Intelligence Imaging Platform In Interstitial Lung Disease Diagnosis, Arjun Prakash Tambe, Ryan Boente, Gautam George, Fayez Kheir, Omid Tahamtani Omran, Kavitha Selvan

Division of Pulmonary, Allergy, and Critical Care Medicine Faculty Papers

Background/Objectives: The diagnosis of interstitial lung disease (ILD) is challenging and frequently delayed. Clinically accessible and minimally invasive diagnostic tools are needed to expedite the diagnosis of ILD while minimizing risk to patients. Fibresolve is an imaging artificial intelligence (AI) tool recently approved by the Food and Drug Administration (FDA) for use in ILD diagnosis and made available to clinicians. The objective of this study was to describe its utility in clinical practice. Methods: We conducted a prospective, observational study of patients across the United States (US) in whom Fibresolve was utilized during routine clinical practice between July 2024 and …