Digital Commons Network^™

A Method For Sensitivity Analysis Of Automatic Contouring Algorithms Across Different Contrast Weightings Using Synthetic Magnetic Resonance Imaging, Lucas Mccullum, Zayne Belal, Warren Floyd, Alaa Mohamed Shawky Ali, Natalie West, Yao Ding, Dan Thill, Nicolette O'Connell, Joseph Stancanello, Kareem A Wahid, David T Fuentes, Ken-Pin Hwang, Clifton D Fuller

Faculty, Staff and Student Publications

Background and purpose: A majority of institution-specific automatic magnetic resonance imaging (MRI)-based contouring algorithms utilize one contrast-weighting (i.e., T2-weighted), however their performance within this contrast-weighting (i.e., across different repetition time, TR, and echo time, TE) is under-investigated and poorly understood. The purpose of this study was to develop a method to evaluate the robustness of automatic contouring algorithms to varying MRI contrast-weightings.

Materials and methods: One healthy volunteer and one patient were scanned using the multi-delay multi-echo (MDME) scan on a 3T MRI. The parotid and submandibular glands in these subjects were contoured using an automatic contouring algorithm trained on …

Semi-Automated Hippocampal Avoidance Whole-Brain Radiotherapy Planning, Dong Joo Rhee, Subha Perni, Kelly J Perrin, Kevin E Casey, Alexandra O Leone, Laurence E Court, Xin Wang, Eun Young Han

Faculty, Staff and Student Publications

Background: Hippocampal avoidance whole-brain radiotherapy (HA-WBRT) is designed to spare cognitive function by reducing radiation dose to the hippocampus during the treatment of brain metastases. Current manual planning methods can be time-consuming and may vary in quality, necessitating the development of automated approaches to streamline the process and ensure consistency.

Purpose: To automate hippocampal avoidance whole-brain radiotherapy (HA-WBRT) planning.

Methods: Our algorithm automatically contours organs-at-risk (OARs) and the hippocampal-avoidance brain target. The algorithm generates planning structures from given contours, utilizing preset beam parameters and dose constraints for optimization. If the dose constraints are unmet, "hotspot" contours will be created to …

Ovarian Cancer Risk And Survival According To Tumor Sex Hormone Receptor Expression: An Ovarian Cancer Association Consortium And Ovarian Tumor Tissue Analysis Consortium Pooled Analysis, Zhuxuan Fu, Lauren Borho, Sarah E Taylor, Linda E Kelemen, Anna Defazio, Martin Köbel, Renhua Na, Antonis C Antoniou, Alison H Brand, Catherine J Kennedy, Nikilyn Nevins, Paul D P Pharoah, Yurii B Shvetsov, Stacey J Winham, Jennifer Alsop, Matthias W Beckmann, Adelyn Bolithon, Jessica Boros, David D L Bowtell, James D Brenton, Michael E Carney, Anita Chudecka-Głaz, Linda S Cook, Cezary Cybulski, Peter A Fasching, Sian Fereday, Renée T Fortner, María J García, Ellen L Goode, Marc T Goodman, Jacek Gronwald, Arndt Hartmann, Brenda Y Hernandez, Estrid Høgdall, David G Huntsman, Allan Jensen, Mercedes Jimenez-Linan, Janine M Joseph, Beth Y Karlan, Ewa Kaznowska, Susanne K Kjaer, Tomasz Kluz, Jennifer M Koziak, Jenny Lester, Teri A Longacre, Maria Lycke, Valerie Mcguire, Kirsten B Moysich, Rachel A Murphy, Sandra Orsulic, Susan J Ramus, Cristina Rodríguez-Antona, Joseph H Rothstein, Spinder Samra, Weiva Sieh, Helen Steed, Karin Sundfeldt, Aline Talhouk, Jan Uciński, Chen Wang, Nicolas Wentzensen, Alice S Whittemore, Lynne R Wilkens, Thomas Songer, Maria Mori Brooks, Lu Tang, Francesmary Modugno

Faculty, Staff and Student Publications

Objective: Many epithelial ovarian cancer (EOC) risk factors relate to sex hormones. The association between these factors and the expression of androgen receptor (AR), estrogen receptor-α (ER), and progesterone receptor (PR) in tumors is unknown.

Method: We linked epidemiologic, AR/ER/PR tumor expression, and survival data from 19 studies in the Ovarian Cancer Association Consortium (OCAC; 4762 cases, 20,888 controls) and the Ovarian Tumor Tissue Analysis (OTTA) consortium (5737 cases). We estimated odds ratios (ORs) and 95 % confidence intervals (CIs) between hormonally-linked factors and tumor AR/ER/PR expression using polytomous logistic regression. We assessed survival by AR/ER/PR tumor expression overall and …

Hyalinization-Based Pathologic Response And Immune Infiltration Following Neoadjuvant Radiotherapy With Or Without Immune-Checkpoint Blockade In Localized Undifferentiated Pleomorphic Sarcoma, R S Traweek, M Zoghbi, R Lazcano, B M Cope, A J Bishop, D Mitra, B A Guadagnolo, D R Ingram, K Wani, D Shamsutdinova, A J Lazar, W-L Wang, C P Scally, E Z Keung, K E Torres, K K Hunt, R Ratan, J A Livingston, M S Nakazawa, D M Araujo, S Patel, V Ravi, A P Conley, M A Zarzour, N Somaiah, C L Roland, E F Nassif Haddad

Faculty, Staff and Student Publications

BACKGROUND: Standard-of-care treatment of localized undifferentiated pleomorphic sarcoma (UPS) involves preoperative radiotherapy (RT) and surgery. Combination of RT with immune-checkpoint blockade (ICB) represents a promising new strategy.

METHODS: Our primary objective was to assess survival outcomes and pathologically documented response in patients receiving RT with or without ICB or chemotherapy.

RESULTS: In a retrospective cohort of 68 patients with primary or locally recurrent UPS who received neoadjuvant RT with or without ICB, the ICB group had greater tumor hyalinization than the non-ICB group (89% compared with 30%, P = 0.018). In the non-ICB group, 19/37 patients (51%) achieved hyalinization-based pathologic …

Dynamic Few-Shot Prompting For Clinical Note Section Classification Using Lightweight, Open-Source Large Language Models, William Hersh, Kirk Roberts, Hongfang Liu

Faculty, Staff and Student Publications

Objective: Unlocking clinical information embedded in clinical notes has been hindered to a significant degree by domain-specific and context-sensitive language. Identification of note sections and structural document elements has been shown to improve information extraction and dependent downstream clinical natural language processing (NLP) tasks and applications. This study investigates the viability of a dynamic example selection prompting method to section classification using lightweight, open-source large language models (LLMs) as a practical solution for real-world healthcare clinical NLP systems.

Materials and methods: We develop a dynamic few-shot prompting approach to classifying sections where section samples are first embedded using a transformer-based …

Hernia Recurrence Risk Matters More Than Mesh Type To Patients Undergoing Ventral Hernia Repair, Stephanie Martinez Ugarte, Mokunfayo O Fajemisin, Chelsea J Guy-Frank, Dalya M Ferguson, Jonah J Stulberg, Lillian S Kao, Julie L Holihan

Faculty, Staff and Student Publications

Introduction: Patients' perceptions about synthetic mesh for ventral hernia repair (VHR) may stem from a lack of knowledge of risks. We assessed patients' threshold for accepting the risk of hernia recurrence to avoid synthetic mesh.

Methods: Patients ≥18 y old with prior or upcoming VHR viewed a one-time video decision aid describing recurrence risk with biologic versus synthetic mesh. Patients chose between these options using a probability trade-off (PTO) technique, varying hernia recurrence rate with biologic mesh. Based on randomized trials, the recurrence rate using synthetic mesh was 9%. The PTO score was the recurrence rate (0%-100%) at which patients …

A Recipe For Disaster - Sodium Bicarbonate Overdose

Faculty, Staff and Student Publications

Audience: Emergency medicine residents post-graduate years 1-4.

Introduction: Sodium bicarbonate is a compound found commonplace in many households and in many products. It lends itself to countless everyday uses including cooking and cleaning. Physicians prescribe sodium bicarbonate regularly both on- and off-label regularly for various ailments.1 Due to the ubiquitous way both emergency physicians and the general public use and are exposed to sodium bicarbonate, we wanted to prepare learners to identify and appropriately manage sodium bicarbonate toxicity.Obtaining a thorough history not just of medications prescribed but of significant ingestions or infusions is critical in diagnosing this toxicity. Furthermore, acute …

Ten Practical Tips For A Successful Presentation At An Infectious Disease Clinical Case Conference

Faculty, Staff and Student Publications

This mini-review provides 10 practical tips for infectious disease trainees presenting at clinical case conferences. These tips, grounded in principles from education theory, aim to enhance the effectiveness of case presentations.

Empirical Tuberculosis Treatment In Human Immunodeficiency Virus (Hiv)-Associated Fever Of Unknown Origin: A Case-Based Rationale

Faculty, Staff and Student Publications

Fever of unknown origin (FUO) in people living with human immunodeficiency virus (PLHIV) is clinically defined as recurrent fever lasting more than four weeks in the outpatient setting or more than three days during hospitalization, despite a thorough diagnostic evaluation. This evaluation typically includes a comprehensive medical history, physical examination, imaging studies (such as chest radiography), and an extensive range of laboratory tests, including complete blood counts, blood and urine cultures, and metabolic panels. Among the many possible causes, tuberculosis (TB) stands out as a leading concern, given its disproportionate burden in PLHIV. However, whether to initiate empirical anti-tubercular therapy …

Cross-Cultural Data On Romantic Love And Mate Preferences From 117,293 Participants Across 175 Countries, Marta Kowal, Piotr Sorokowski, Biljana Gjoneska, Katarzyna Pisanski, Gerit Pfuhl, Leonardo Aguilar, Steve M.J. Janssen, Benjamin Gelbart, Patrícia Arriaga, Jan Antfolk, Katarina Zvončáková, Linda H. Lidborg, Jorge Contreras-Garduño, Mikhail V. Kozlov, Taciano L. Milfont, Marco A.C. Varella, Valerija Križanić, Mahmoud Boussena, Tina Kavčič, Diana Ribeiro Da Silva, Brahim Hamdaoui, Fatima Zahra Sahli, Karlijn Massar, Eliane Deschrijver, Tatsunori Ishii, Hakan Cetinkaya, Oksana Senyk, Farida Guemaz, Koen Ponnet, Yahya Don, Dušana Šakan, Gyesook Yoo, Ravit Nussinson, Joaquín Ungaretti, Ali R. Can, Izzet Duyar, Jiří Čeněk, Joao Carneiro, Norbert Meskó, Luca Kozma, Ellen K. Nyhus, Mona Vintila, Oulmann Zerhouni, Farid Pazhoohi, Maja Zupančič, Sinem Söylemez, Austin H.-E. Wang, Marietta Papadatou-Pastou, Irena Pavela Banai, Pavol Prokop, Mohd Sofian Omar Fauzee, Reza Afhami, Jean C. Natividade, Roberto Baiocco, Mara Morelli, Toivo Aavik, Ezgi Toplu-Demirtaş, Singha Tulyakul, Anna Wlodarczyk, Razleh Chegeni, Anabela C. Santos, Dmitry Grigoryev, Dmitrii Dubrov, Dimitri Chubinidze, Gözde Ikizer, Nana Burduli, Johanna Czamanski-Cohen, Rizwana Amin, Petros Roussos, Evgeniya Hristova, Rūta Sargautytė, Ekaterine Pirtskhalava, Tenuunjargal Avirmed, Arooj Najmussaqib, Abdelilah Charyate, Shagufta Batool, Tatiana Volkodav, Yoshihiko Kunisato, Yuki Yamada, Asako Toyama, Mariia Perun, Seda Dural, Tetyana Mandzyk, Anna Studzinska, Ognen Spasovski, Felipe E. García, Caterina Grano, Merve Boğa, Mehmet Koyuncu, Sangeeta Singh, Ju Hee Park, Derya Atamtürk, Samuel Lins, Martin Pírko, David Lacko, Balazs Aczel, Ferenc Kocsor, Ádám Putz, Tobias Otterbring, Pavol Kačmár, Efisio Manunta, Théo Besson, Nasim Ghahraman Moharrampour, Çağlar Solak, Ignacio Estevan, Merve Topcu Bulut, Nicolas Kervyn, Moises Mebarak, Jackson G. Lu, Nejc Plohl, Bojan Musil, Adil Samekin, Kirill G. Miroshnik, Clément Cornec, Isabella Giammusso, Ulf-Dietrich Reips, Maria Rosa Miccoli, Miriam Parise, Sabrina Stöckli, Tiago Marot, Sibele D. Aquino, Amanda Londero-Santos, Antonio Chirumbolo, Aybegum Memisoglu-Sanli, Jaroslava V. Valentova, Cemre Karaarslan, Ivana Hromatko, Kevin S. Kertechian, Ogeday Çoker, Matheus F. F. Ribeiro, Carlota Batres, Ilker Dalgar, Stephanie J. Eder, Katarina Mišetić, Marios Argyrides, Vita Mikuličiūtė, Silvia Mari, Elisabeth Oberzaucher, Kathrin Masuch, Alan D. A. Mattiassi, Salma Samir Omar, Elena Piccinelli, Eda Ermagan Caglar, Diogo Lamela, David A. Frederick, Aleksander Kobylarek, Ma. Criselda T. Pacquing, Marc Eric Santos Reyes, Marcos Zumárraga-Espinosa, Feten Fekih-Romdhane, Talía Gómez Yepes, Edgardo Etchezahar, Katarzyna Galasinska, Jan P. Röer, Ayşegül Şahin, Miguel Landa-Blanco, Arkadiusz Urbanek, Chee-Seng Tan, Rita Castro, Ksenija Cunichina, Anna Krasnodębska, Daniel Conroy-Beam, Franciszek Ostaszewski, Izabela Chałatkiewicz, Beatriz Abad-Villaverde, Bastien Trémolière, Alexios Arvanitis, Gulmira Topanova, Wiliam J. Chopik, Grace Akello, Ariela F. Pagani, Silvia Donato, Peter Fedor, Tomasz Frackowiak, Simon Ozer, Marlon Mayorga-Lascano, Farah Khan, Maryanne L. Fisher, Princess Lovella G. Maturan, Tatiana Semenovskikh, Sanjana Dutt, William Tamayo-Agudelo, Gulnara Ismukhanova, Laith Al-Shawaf, Luisa Angelucci, Adam Bode, Sercan Balım, Jovi C. Dacanay, Chiemezie S. Atama, Kai A. D. Morgan Campbell, Tchilissila A. Simões, Barış Özener, Paula Błauciak, Filipe Prazeres

Psychology Faculty Articles and Research

Psychological studies on close relationships have often overlooked cultural diversity, dynamic processes, and potentially universal principles that shape intimate partnerships. To address the limited generalizability of previous research and advance our understanding of romantic love experiences, mate preferences, and physical attractiveness, we conducted a large-scale cross-cultural survey study on these topics. A total of 404 researchers collected data in 45 languages from April to August 2021, involving 117,293 participants from 175 countries. Aside from standard demographic questions, the survey included valuable information on variables relevant to romantic relationships: intimate, passionate, and committed love within romantic relationships, physical-attractiveness enhancing behaviors, gender …

Visualizing Seascapes As A Method For Engaging Stakeholders In Discussions About Resilience, Dominica Williamson, Louisa Evans, Mark Bryant, Caroline Hattam, Tara Hooper, Andy Hughes, Timur Jack-Kadioglu, Fazlun Khalid, John Martin, Celine Miternique, Karyn Morrissey, Lorna Slade, Ali Thani, Kathy Young

Professional Services

Coastal communities have complex bodies of knowledge around environmental change, resilience, and sustainability, which extractive forms of research are not well suited to understand. This paper outlines an arts-based action research pilot project to understand social-ecological resilience in the Western Indian Ocean. Our project used visual, artistic, and participatory approaches to foster co-learning and action research on: (1) the arts-based methods themselves and (2) resilience building in coastal communities and ecosystems. This paper focuses on the former elements. We present our inherently phenomenological approach as a research assemblage that can be used to bring together different communities and stakeholders, disciplines, …

Aesthetic Judgment: Considering A Contemporary Standard Of Taste, Lydia Margaret Lee

Theses and Dissertations

Originally published in 1757 as part of his Four Dissertations essay collection, David Hume’s essay, “Of the Standard of Taste,” occupies a prominent place in Enlightenment aesthetic studies. Due to an oversaturation of print materials being produced, eighteenth century society was concerned with evaluating and ranking the massive quantities of art being produced. Philosophers such as David Hume attempted to define criteria for determining a standard of taste, aiming to hierarchically rank works of art based on aesthetic experience. For centuries, critics such as Timothy Costelloe, Noel Carroll, David Marshall, Jonathan Friday, James Shelley, and many others have engaged in …

Improving Professional Learning Communities (Plcs) For Singletons, Stephanie Harbulak-Barron

Theses and Dissertations

Participation in professional learning is generally recognized as a required component of the teaching profession. One type of professional learning is professional learning communities (PLCs). However, there are challenges to implementing successful PLCs that result in relevant professional learning. One challenge is non-standard PLC groupings that contain singletons, especially when those PLC groups are primarily all singletons. This action research study examined educators’ and administrators’ expectations of and experiences with PLCs in a special education cooperative in the Midwest. The goal of the mixed methods research design was to gather enough information from participants to facilitate the creation of an …

Mtu-Llm: Llm-Based Multi-Robot Task Allocation And Path Planning For Heterogeneous Robots In Search And Rescue Operations, Kaushik Kannan, Jungyun Bae

Michigan Tech Publications

Urban Search and Rescue operations after natural disasters involve locating and assisting victims in hazardous environments, which is challenging. Classical Multi-Robot Task Allocation (MRTA) and path planning approaches have been used to deploy heterogeneous robot teams in unsafe areas. However, existing methods often lack focus on workload balance and requirement fulfillment and struggle to generalize across different scenarios. To address these challenges, we propose Multi-robot Task allocation Utilizing LLMs (MTU-LLM), a framework designed to reduce the development time for task allocation and path planning approaches, enabling faster robot deployment. The framework uses an LLM-based “prompt engineering” approach that generates task …

Performance Characterization And Mechanistic Study Of Dry Process Wr/Sbs Composite Modified Asphalt Mixture, Hongfu Liu, Xiangyang Ren, Wenwen Ge, Zhengwei Yi, Dongzhao Jin, Xinghai Peng, Songtao Lv

Michigan Tech Publications

The dry process modification is characterized by simplicity, flexibility, low energy consumption, and low pollution, making it an ideal choice in pavement repair engineering. This study aims to optimize the use of waste rubber powder and SBS modifiers in dry process composite modification by preparing dry process WR/SBS composite modified particles (WR/SBS CMP) using a twin-screw extruder. The optimum mixing content of modified particles was determined through comparative analysis of pavement performance of dry process WR/SBS composite modified, wet-process SBS-modified, and matrix asphalt mixtures. The rheological properties and modification mechanism of the asphalt binder extracted from these mixtures were tested …

Machine Learning For Parkinson’S Disease: A Comprehensive Review Of Datasets, Algorithms, And Challenges, Sahar Shokrpour, Amir Mehdi Moghadamfarid, Sepideh Bazzaz Abkenar, Mostafa Haghi Kashani, Mohammad Akbari, Mostafa Sarvizadeh

Michigan Tech Publications

Parkinson’s disease (PD) is a devastating neurological ailment affecting both mobility and cognitive function, posing considerable problems to the health of the elderly across the world. The absence of a conclusive treatment underscores the requirement to investigate cutting-edge diagnostic techniques to improve patient outcomes. Machine learning (ML) has the potential to revolutionize PD detection by applying large repositories of structured data to enhance diagnostic accuracy. 133 papers published between 2021 and April 2024 were reviewed using a systematic literature review (SLR) methodology, and subsequently classified into five categories: acoustic data, biomarkers, medical imaging, movement data, and multimodal datasets. This comprehensive …

Stability Analysis Of The Eulerian-Lagrangian Finite Volume Methods For Nonlinear Hyperbolic Equations In One Space Dimension, Yang Yang, Jiajie Chen, Jing Mei Qiu

Michigan Tech Publications

In this paper, we construct a novel Eulerian-Lagrangian finite volume (ELFV) method for nonlinear scalar hyperbolic equations in one space dimension. It is well known that the exact solutions to such problems may contain shocks though the initial conditions are smooth, and direct numerical methods may suffer from restricted time step sizes. To relieve the restriction, we propose an ELFV method, where the space-time domain was separated by the partition lines originated from the cell interfaces whose slopes are obtained following the Rakine-Hugoniot junmp condition. Unfortunately, to avoid the intersection of the partition lines, the time step sizes are still …

The B2biers System: A Content-Based Perspective On Maximizing Influence And Subscription In Social Networks, Konstantinos Theocharidis, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

The popular problem of Influence Maximization (IM) asks for the k users who can maximize the influence of a fixed post in a social network. In contrast, the problem of Content- Aware Influence Maximization (CAIM) asks for the k features to form a viral tunable post in a social network starting its diffusion from a fixed set of initial adopters. CAIM paves the way for a number of novel problems to be studied that altogether can lead to the development of a system that would be valuable for advertisers who manage social network pages. This holds since features (brands) in …

Evaluating Chatgpt To Answer Multi-Modal Exercises In Computer Science Education, Eng Lieh Ouh, Kar Way Tan, Siaw Ling Lo, Benjamin Gan

Research Collection School Of Computing and Information Systems

This study investigates ChatGPT-4o's ability to answer multi-modal assessment exercises in computer science (CS) courses. While the use of large language models (LLMs) to answer text-based exercises are extensively researched, their ability to answer exercises involving artifacts of other modalities remains underexplored. To close this gap, we evaluate ChatGPT-4o's answers to 120 multi-modal CS exercises in programming, software design, human-computer interaction, statistical analysis, process analysis, and simulation. The multi-modal artifacts in these exercises include class diagrams, sequence diagrams, user interface images, analytical charts, workflow diagrams and object-flow diagrams. Our comparisons to the expected answers of these exercises show that ChatGPT-4o …

Crow: Eliminating Backdoors From Large Language Models Via Internal Consistency Regularization, Nay Myat Min, Long H. Pham, Yige Li, Jun Sun

Research Collection School Of Computing and Information Systems

Large Language Models (LLMs) are vulnerable to backdoor attacks that manipulate outputs via hidden triggers. Existing defense methods—designed for vision/text classification tasks—fail for text generation. We propose Internal Consistency Regularization (CROW), a defense leveraging the observation that backdoored models exhibit unstable layer-wise hidden representations when triggered, while clean models show smooth transitions. CROW enforces consistency across layers via adversarial perturbations and regularization during finetuning, neutralizing backdoors without requiring clean reference models or trigger knowledge—only a small clean dataset. Experiments across Llama-2 (7B, 13B), CodeLlama (7B, 13B), and Mistral-7B demonstrate CROW’s effectiveness: it achieves significant reductions in attack success rates across …

An On-The-Fly Synthesis Framework For Ltl Over Finite Traces, Shengping Xiao, Yongkang Li, Shufang Zhu, Jun Sun, Jianwen Li, Geguang Pu, Moshe Vardi

Research Collection School Of Computing and Information Systems

We present an on-the-fly synthesis framework for Linear Temporal Logic over finite traces (LTLf) based on top-down deterministic automata construction. Existing approaches rely on constructing a complete Deterministic Finite Automaton (DFA) corresponding to the LTLf specification, a process with doubly exponential complexity relative to formula size in the worst case. In this case, the synthesis cannot be conducted until the entire DFA is constructed. This inefficiency is the main bottleneck of existing approaches. To address this challenge, we first present a method for converting LTLf into Transition-based DFA (TDFA) by directly leveraging LTLf semantics, incorporating intermediate results as direct components …

Efficient Prompt Tuning For Hierarchical Ingredient Recognition, Yinxuan Gui, Bin Zhu, Jingjing Chen, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

Fine-grained ingredient recognition presents a significant challenge due to the diverse appearances of ingredients, resulting from different cutting and cooking methods. While existing approaches have shown promising results, they still require extensive training costs and focus solely on fine-grained ingredient recognition. In this paper, we address these limitations by introducing an efficient prompt-tuning framework that adapts pretrained visual-language models (VLMs), such as CLIP, to the ingredient recognition task without requiring full model finetuning. Additionally, we introduce three-level ingredient hierarchies to enhance both training performance and evaluation robustness. Specifically, we propose a hierarchical ingredient recognition task, designed to evaluate model performance …

Dual-Target Disjointed Cross-Domain Recommendation Mediated Via Latent User Preferences, Dinh Hieu Do, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

Users often navigate multiple platforms online, each characterized by its own set of scarce data. Recommender systems face a significant challenge in such fragmented environments. This paper proposes a novel approach to enhance recommendation systems by leveraging connections across distinct yet conceptually similar datasets from multiple platforms. We introduce a unique scenario of dual-target overlapping-free cross-platform recommendation, presenting a bridging mechanism to mutually improve across platforms and learn latent user preferences. Our approach addresses the data sparsity prevalent in each platform and enhances recommendation quality by harnessing redundant, rich, and similar domain data. Experiments validate the effectiveness of our method, …

Advancing Food Nutrition Estimation Via Visual-Ingredient Feature Fusion, Huiyan Qi, Bin Zhu, Chong-Wah Ngo, Jingjing Chen, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Nutrition estimation is an important component of promoting healthy eating and mitigating diet-related health risks. Despite advances in tasks such as food classification and ingredient recognition, progress in nutrition estimation is limited due to the lack of datasets with nutritional annotations. To address this issue, we introduce FastFood, a dataset with 84,446 images across 908 fast food categories, featuring ingredient and nutritional annotations. In addition, we propose a new model-agnostic Visual-Ingredient Feature Fusion (VIF2 ) method to enhance nutrition estimation by integrating visual and ingredient features. Ingredient robustness is improved through synonym replacement and resampling strategies during training. The ingredient-aware …

Hps: Hard Preference Sampling For Human Preference Alignment, Xiandong Zou, Wanyu Lin, Yuchen Li, Pan Zhou

Research Collection School Of Computing and Information Systems

Aligning Large Language Model (LLM) responses with human preferences is vital for building safe and controllable AI systems. While preference optimization methods based on PlackettLuce (PL) and Bradley-Terry (BT) models have shown promise, they face challenges such as poor handling of harmful content, inefficient use of dispreferred responses, and, specifically for PL, high computational costs. To address these issues, we propose Hard Preference Sampling (HPS), a novel framework for robust and efficient human preference alignment. HPS introduces a training loss that prioritizes the most preferred response while rejecting all dispreferred and harmful ones. It emphasizes “hard” dispreferred responses — those …

Repairing Adversarial Texts Through Perturbation, Guoliang Dong, Jingyi Wang, Jun Sun, Sudipta Chattopadhyay, Xinyu Wang, Ting Dai, Jie Shi, Jin Song Dong

Research Collection School Of Computing and Information Systems

It is known that neural networks are subject to attacks through adversarial perturbations. Worse yet, such attacks are impossible to eliminate, i.e., the adversarial perturbation is still possible after applying mitigation methods such as adversarial training. Multiple approaches have been developed to detect and reject such adversarial inputs. Rejecting suspicious inputs however may not be always feasible or ideal. First, normal inputs may be rejected due to false alarms generated by the detection algorithm. Second, denial-of-service attacks may be conducted by feeding such systems with adversarial inputs. To address this, in this work, we focus on the text domain and …

Unbounded Multi-Hop Proxy Re-Encryption With Hra Security: An Lwe-Based Optimization, Xiaohan Wan, Yang Wang, Haiyang Xue, Mingqiang Wang

Research Collection School Of Computing and Information Systems

Proxy re-encryption (PRE) schemes enable a semi-honest proxy to transform a ciphertext of one user i to another user j while preserving the privacy of the underlying message. Multi-hop PRE schemes allow a legal ciphertext to undergo multiple transformations, but for lattice-based multi-hop PREs, the number of transformations is typically bounded due to the increase of error terms. Recently, Zhao et al. (ESORICS 2024) introduced a lattice-based unbounded multi-hop (homomorphic) PRE scheme that supports an unbounded number of hops. Nevertheless, their scheme only achieves the selective CPA security. In contrast, Fuchsbauer et al. (PKC 2019) proposed a generic framework for …

An Incentive Mechanism For Privacy Preserved Data Trading With Verifiable Data Disturbance, Man Zhang, Xinghua Li, Bin Luo, Yanbing Ren, Yinbin Miao, Ximeng Liu, Robert H. Deng

Research Collection School Of Computing and Information Systems

To motivate data owners’ (DOs’) trading willingness, the existing incentive mechanisms allow DOs to independently disturb data following data consumer's (DC’s) availability requirement. However, they cannot motivate DOs’ honest disturbance, which is attributed to DOs’ independent disturbance without any supervision. Thus, we implement an incentive mechanism for privacy preserved data trading with verifiable data disturbance where an honest-but-curious disturbance generator (DG) is additionally introduced to supervise DOs’ local disturbance and assist disturbance verification between DOs and DC. Specifically, DG generates the disturbance strategies and secretly distributes to DOs following private information retrieval, guaranteeing DOs's local disturbance's privacy and verifiability with …

Sanitizable Cross-Domain Access Control With Policy-Driven Dynamic Authorization, Jianfei Sun, Guowen Xu, Hongwei Li, Tianwei Zhang, Cong Wu, Xuehuan Yang, Robert H. Deng

Research Collection School Of Computing and Information Systems

The increasing demand for secure and efficient data sharing has underscored the importance of developing robust cryptographic schemes. However, many existing endeavors have overlooked the following critical issues: (1) unauthorized access resulting from malicious information leakage by senders; (2) absence of constraints on write and read permissions for participants; (3) and inflexibility of strategies to dynamically designate ciphertexts to multiple recipients. In this paper, we present SCPA, a cross-domain access control scheme imbued with sanitization features and propelled by policy-driven dynamic authorization, tailored for cloud-based data sharing. This scheme not only facilitates access controls, including regulations for no-read and no-write …

Understanding The Bad Development Practices Of Android Custom Permissions In The Wild, Xiaohan Zhang, Zhiyuan Yu, Xinghua Li, Cen Zhang, Cong Sun, Ning Zhang, Robert H. Deng

Research Collection School Of Computing and Information Systems

Android system provides application developers with the ability to define custom permissions, which serve to moderate the sharing of resources and interactions with other applications. However, poor development practices of developers can render the permission mechanism ineffective, weakening the system protection. This paper presents a comprehensive examination of the problematic practices surrounding custom permissions employed by developers, referred to as Bad Practices of Custom Permissions (BPCP issues). To accomplish this, we conducted an empirical study and identified nine common BPCP issue patterns that can lead to various adverse consequences, such as installation failures, crashes, or even component hijacking. To automatically …