Artificial Intelligence and Robotics Commons^™

Codeultrafeedback: An Llm-As-A-Judge Dataset For Aligning Large Language Models To Coding Preferences, Martin Weyssow, Aton Kamanda, Xin Zhou, Houari Sahraoui

Research Collection School Of Computing and Information Systems

Evaluating the alignment of large language models (LLMs) with user-defined coding preferences is a challenging endeavor that requires a deep assessment of LLMs' outputs. Existing methods and benchmarks rely primarily on automated metrics and static analysis tools, which often fail to capture the nuances of user instructions and LLM outputs. To address this gap, we introduce the LLM-as-a-Judge evaluation framework and present CodeUltraFeedback, a comprehensive dataset for assessing and improving LLM alignment with coding preferences. CodeUltraFeedback consists of 10,000 coding instructions, each annotated with four responses generated from a diverse pool of 14 LLMs. These responses are annotated using GPT-3.5 …

Learning To Search For Vehicle Routing With Multiple Time Windows, Kuan Xu, Zhiguang Cao, Chenlong Zheng, Lindong Liu

Research Collection School Of Computing and Information Systems

In this study, we propose a reinforcement learning-based adaptive variable neighborhood search (RL-AVNS) method designed for effectively solving the Vehicle Routing Problem with Multiple Time Windows (VRPMTW). Unlike traditional adaptive approaches that rely solely on historical operator performance, our method integrates a reinforcement learning framework to dynamically select neighborhood operators based on real-time solution states and learned experience. We introduce a fitness metric that quantifies customers’ temporal flexibility to improve the shaking phase, and employ a transformer-based neural policy network to intelligently guide operator selection during the local search. Extensive computational experiments are conducted on realistic scenarios derived from the …

Compositions Of Variant Experts For Integrating Short-Term And Long-Term Preferences, Dinh Hieu Do, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

In the online digital realm, recommendation systems are ubiquitous and play a crucial role in enhancing user experience. These systems leverage user preferences to provide personalized recommendations, thereby helping users navigate through the paradox of choice. This work focuses on personalized sequential recommendation, where the system considers not only a user’s immediate, evolving session context, but also their cumulative historical behavior to provide highly relevant and timely recommendations. Through an empirical study conducted on diverse real-world datasets, we have observed and quantified the existence and impact of both short-term (immediate and transient) and long-term (enduring and stable) preferences on users’ …

Efficient Active Training For Deep Lidar Odometry, Beibei Zhou, Zhiyuan Zhang, Zhenbo Song, Jianhui Guo, Hui Kong

Research Collection School Of Computing and Information Systems

Robust and efficient deep LiDAR odometry models are crucial for accurate localization and 3D reconstruction, but typically require extensive and diverse training data to adapt to diverse environments, leading to inefficiencies. To tackle this, we introduce an active training framework designed to selectively extract training data from diverse environments, thereby reducing the training load and enhancing model generalization. Our framework is based on two key strategies: Initial Training Set Selection (ITSS) and Active Incremental Selection (AIS). ITSS begins by breaking down motion sequences from general weather into nodes and edges for detailed trajectory analysis, prioritizing diverse sequences to form a …

Identifying And Mitigating Api Misuse In Large Language Models, Terry Yue Zhuo, Junda He, Jiamou Sun, Zhenchang Xing, David Lo, John Grundy, Xiaoning Du

Research Collection School Of Computing and Information Systems

API misuse in code generated by large language models (LLMs) presents a serious and growing challenge in software development. While LLMs demonstrate impressive code generation capabilities, their interactions with complex library APIs are often error-prone, potentially leading to software failures and vulnerabilities. In this paper, we conduct a large-scale study of API misuse patterns in LLM-generated code, analyzing both method selection and parameter usage across Python and Java, using three representative LLMs (StarCoder-7B, Qwen2.5-Coder-7B, and GitHub Copilot). Based on extensive manual annotation of 3,209 method-level and 3,492 parameter-level misuses, we identify and categorize four recurring misuse types by building on …

Addressing Graph Heterogeneity And Heterophily From A Spectral Perspective, Kangkang Lu, Yanhua Yu, Ruopei Guo, Nan Cheng, Zhiyong Huang, Yunshan Ma, Meiyu Liang, Yuling Wang, Xiting Qin, Yimeng Ren, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Graph Neural Networks (GNNs) face two key challenges, heterogeneity and heterophily, which often degrade performance. Existing approaches either focus narrowly on specific meta-paths, limiting their expressiveness, or are expressive but cannot effectively leverage higher-order neighbors. In this paper, we propose the Heterogeneous Heterophilic Spectral Graph Neural Network (H2SGNN), which combines local independent filtering to adaptively handle meta-path subgraphs with varying homophily ratios, and global hybrid filtering to capture high-order neighbor interactions with linear computational complexity. On five heterogeneous graph benchmarks—DBLP, ACM, IMDB, AMiner, and Yelp—H2SGNN consistently outperforms strong baselines, for example, achieving +1.0% Macro-F1 and +1.3% Micro-F1 on IMDB. It …

Comprehensively Evaluating The Perception Systems Of Autonomous Vehicles Against Hazards, Xiaodong Zhang, Jie Bao, Jianlei Chi, Jun Sun, Zijiang Yang

Research Collection School Of Computing and Information Systems

Perception systems are vital for the safety of autonomous driving. In complex autonomous driving scenarios, autonomous vehicles must overcome various natural hazards, such as heavy rain or raindrops on the camera lens. Therefore, it is essential to conduct comprehensive testing of the perception systems in autonomous vehicles against these hazards, as demanded by the regulatory agencies of many countries for human drivers. Since there are many hazard scenarios, each of which has multiple configurable parameters, the challenges are (1) how do we systematically and adequately test an autonomous vehicle against these hazard scenarios, with measurable outcome; and (2) how do …

Interpretable Machine Learning For Personalized Profiling Of Mild Cognitive Impairment From Daily Activities, Budhitama Subagdja, Ah-Hwee Tan, Kenneth Kwok, Iris Rawtaer

Research Collection School Of Computing and Information Systems

Continuous monitoring of individual daily activities is essential to detect mild cognitive impairment (MCI) wherein timely intervention can still be applied to prevent more severe mental decline. Recent approaches in predicting MCI are mostly considering digital biomarkers across individuals but often neglecting specific indicators from a single person over a long period of time. Making this personalized, dynamic, and highly noisy prediction model with irregular distribution of missing information to be explainable and actionable for clinical use, remains a challenge. This paper presents a study on a personalized MCI prediction and profiling from an in-home and mobile cognitive health monitoring …

Hybrid Deep Learning For Anti-Money Laundering: Unsupervised Detection Of Emerging Schemes Via Feature Fusion And Explainable Artificial Intelligence, Cosmas Ochieng Kungu, Kennedy Senagi, Evans Omondi

All Peer-Reviewed Publications

Traditional rule-based anti-money laundering (AML) transaction monitoring systems suffer from high false-positive rates and rigidity in detecting complex emerging risk. This limitation has prompted changes to the Financial Action Task Force (FATF) recommendation 16, mandating the use of advanced systems for detecting money laundering schemes in cross-border payments. This study developed a hybrid framework integrating VAE-learned behavioural latent factors, GNN-captured relational network signals, and rule-based heuristics for enhanced anomaly detection. The model was evaluated on 54,258 real-world cross-border transaction records from an East African commercial bank. The One-Class SVM, optimised via a rigorous grid search proved superior compared to Isolation …

Automated Machine Learning For High-Resolution Daily And Hourly Methane Emission Mapping For Rice Paddies Over South Korea: Integrating Modis, Era5-Land, And Soil Data, Jiah Jang, Seung Hee Kim, Menas Kafatos, Jaeil Cho, Gayoung Yoo, Sujong Jeong, Yangwon Lee

Institute for ECHO Articles and Research

Agriculture is a major global source of methane (CH4), and accurate emission estimates are essential for refining national greenhouse gas inventories and supporting climate-resilient policies. This study develops a high-resolution estimation framework for CH4 emissions from Korean rice paddies by integrating multi-source datasets, including Moderate Resolution Imaging Spectroradiometer (MODIS) vegetation indices, European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis Version 5 (ERA5)-Land meteorological variables, and Harmonized World Soil Database (HWSD) soil properties. Using CH4 flux observations from four global rice ecosystems (Italy, Japan, South Korea, and USA), we constructed parallel daily and hourly machine learning models using an automated machine …

How Agile Became The Design Philosophy Of Ai Fishbowl Under Real-World Constraints, Jad Saad

University Honors Theses

This capstone review examines the development of AI Fishbowl, a public-facing, interactive artificial intelligence system, as a case study in how Agile methods evolve from a project management tool into a design philosophy under real-world constraints. Although the project adopted an Agile workflow early on through a Kanban-style task management approach, the initial system design and architecture were still shaped by a largely plan-first mindset. This created a mismatch between flexible process and rigid design assumptions, which became increasingly apparent as the team moved from high-level architecture into implementation.

A critical turning point occurred when early architectural plans proved difficult …

A High-Fidelity Multimodal Synthetic Dataset Generation Framework For Off-Road Unstructured Terrain Navigation Training Of Autonomous Robots, Liyana Wijayathunga, Dulitha Dabare, Alexander Rassau, Douglas Chai, Syed Mohammed Shamsul Islam

Research outputs 2022 to 2026

The success of deep learning methods in a wide range of application areas has inspired many recent developments in the urban and off-road autonomous navigation domain. In particular, techniques for semantic scene understanding, a key aspect of the navigation pipeline, have been researched extensively, resulting in many real-world and synthetic datasets. However, in comparison to urban semantic segmentation datasets, the availability of datasets for off-road environments remains sparse. In this paper, we aim to overcome this challenge by introducing a methodology capable of efficiently generating photorealistic synthetic datasets for off-road environments with support for multiple sensor modalities. The developed approach …

Using Large Language Models To Analyze Political Texts Through Natural Language Understanding, Kenneth Benoit, Scott De Marchi, Conor Laver, Michael Laver, Jinshuai Ma

Research Collection School of Social Sciences

Large language models (LLMs) offer scalable alternatives to human experts when analyzing political texts for meaning, using natural language understanding (NLU). Qualitative NLU methods relying on human experts are severely limited by cost and scalability. Statistical text-as-data methods are scalable but rely on strong and often unrealistic assumptions. We propose a systematic, scalable, and replicable method that can extend existing qualitative and quantitative approaches by using LLMs to interpret texts meaningfully rather than as mere data. Our ensemble means of LLM-generated estimates of party positions on six key issue dimensions correlate highly with equivalent mean ratings by country specialists. When …

Algorithmic Opacity In Opioid Risk Scoring And The Need For Transparent Ai Regulation, Sherry Yun Wang, Ryan Stofer, Zhouzhou Chu, Xiao Huang, Ang Li

Pharmacy Faculty Articles and Research

NarxCare®, a proprietary opioid risk scoring system embedded in Prescription Drug Monitoring Programs (PDMPs), has generated significant patient complaints. We adhered to the technical specifications and applied them to PDMP and IQVIA PharMetrics® Plus Closed Health Plan claims database. Despite adding socioeconomic covariates, precision (0.01–0.32) was far below the reported benchmark of 0.75, and F1 scores (0.02–0.39) were also substantially lower than the benchmark value of 0.65, across all our reconstructed models.

Artificial Intelligence-Driven Materials Science: Evolution, Framework, Dilemmas, And Breakthroughs, Yanglili Zhou, Weihua Wang, Ziwei Zhao

Bulletin of Chinese Academy of Sciences (Chinese Version)

Artificial intelligence-driven materials science (AIMS) represents a revolutionary and disruptive paradigm in materials research, promising to fundamentally break through the traditional bottlenecks of research cycles and efficiency. Historically, the evolution of materials science research paradigms from empirical trial and error, theoretical modeling, and computational simulation to the new data-driven stage has been driven by innovations in cognitive tools and methods. Currently, artificial intelligence, as a disruptive cognitive tool, is fundamentally reconstructing the core elements and interaction logic of materials science: the research process achieves intelligent iteration and full-process closed-loop; the capabilities of researchers are reshaped and teams are organized; and …

Study On Strategies And Measures And Insights For Introducing Artificial Intelligence Talents In Japan, Fangxin Hu, Ru Ma, Yujun Su

Bulletin of Chinese Academy of Sciences (Chinese Version)

In response to the aging population and the need for innovation in science and technology development, Japan regards AI as a key technology to solve social problems. In addition to accelerating the training of domestic AI talents, Japan is also vigorously introducing overseas AI talents. This study sorts out and analyzes Japan’s long-term, annual, and AI-specific strategic planning for the introduction of AI talents, including Basic Plan for Science, Technology and Innovation, Comprehensive Innovation Strategy, Strategic Plan for Artificial Intelligence Technology, and AI Strategy, and explores Japan’s specific implementation measures such as updating the national residence management system, improving the …

Cerebral Documents And Algorithmic Sensemaking: Searching For Expressions In Human And Artificial Cognitive Collaborations, Rebekah L. Cowell

Proceedings from the Document Academy

Generative Artificial Intelligences (AIs) and current advanced large language models (LLMs) are algorithmically designed to generate text-based conversations as conversational agents (CAs), by replicating human language and conversational communication. Pairing human cognition with generative computationally coded cognition. We have never been here before: cerebral and artificial information collaborations and processing producing expressions that may or may not become visible as second-hand/secondary source documents.

Sensemaking or sense(un)making is a unique autonomous human drive cognitively, our information processing is sensemaking in action and expressions and articulations are evidence of the sensemaking cycle. Documentation [expressed or articulated through various mediums] are a product …

Ueof, Nick Truong, Pritam P. Karkomar, William J. Beksi

Event-Based Vision - Archive

UEOF is the first synthetic underwater event-based optical flow dataset derived from physically-based ray-traced RGBD sequences. It was constructed using a modern video-to-event pipeline applied to rendered underwater videos. It consists of realistic event data streams with dense ground-truth flow, depth, and camera motion. The dataset is composed of 12 minutes and 51 seconds of data across 13,714 RGB frames. This results in a total of 4.94 billion events across all scenes. UEOF exhibits a high dynamic range of motion with a mean flow magnitude of 6.1 px and a median of 3.6 px. The motion distribution is heavy-tailed. While …

Ai-Guided Binding Mechanisms And Molecular Dynamics For Mers-Cov, Pradyumna Kumar, Lingtao Chen, Rachel Yuanbao Chen, Yin Chen, Seyedamin Pouriyeh, Progyateg Chakma, Abdur Rahman Mohd Abul Basher, Yixin Xie

Faculty Articles

The MERS-CoV (Middle East respiratory syndrome coronavirus) is a zoonotic virus with a high mortality rate and a lack of antiviral drugs, underscoring the need for effective therapeutic methods. Viral entry depends on interactions between viral surface proteins and human receptors, with Dipeptidyl Peptidase-4 (DPP4), a transmembrane glycoprotein, acting as the receptor for MERS-CoV. We employed Molecular Dynamics (MD) Simulations to identify critical interface residues under a high-performance computing (HPC) workflow for accelerated results. Target residue pairs were identified through analysis of salt bridge and hydrogen bond occupancy. The stability of these residues was confirmed through three independent MD Simulations …

Research On System And Application Framework Of Tactical Wargaming Simulation Driven By Ai4s, Dayong Liu, Qisheng Guo, Zhiming Dong, Xuehuan Qiu, Zhuoli Liu

Journal of System Simulation

Abstract: Tactical wargaming simulation, as a crucial tool for combat analysis, simulation training, and equipment demonstration and test, has become a significant means for generating combat effectiveness. Integrating AI into simulation not only enhances simulation efficiency but also diminishes reliance on humans. To assist professionals engaged in tactical wargaming simulation in mastering AI application methods, fostering a systematic mindset, and understanding evolving trends, this paper provided a concise overview of the principles behind AI for science (AI4S). Subsequently, it conducted an analysis of AI4S's application effectiveness in tactical wargaming simulation, established an AI4S-driven wargaming simulation system, and elucidated its composition, …