Digital Commons Network^™

Operationalizing Ethics For Ai Agents: How Developers Encode Values Into Repository Context Files, Christoph Treude, Sebastian Baltes, Marc Cheong

Research Collection School Of Computing and Information Systems

As AI coding agents become embedded in software development workflows, developers are beginning to operationalize ethical principles by encoding behavioral rules into repository-level context files for AI agents, such as AGENTS.md files. Rather than examining the ethics of AI agents in the abstract, this vision paper investigates how ethics and values are already being translated for AI agents into actionable instructions that shape agent behavior. Through a preliminary investigation, we find that developers are already embedding guidance related to fairness, accessibility, sustainability, tone, and privacy. These artifacts function as a developer-authored governance layer, translating abstract principles into situated, natural-language directives …

A Dataset Of Agentic Ai Coding Tool Configurations, Matthias Galster, Seyedmoein Mohsenimofidi, Levi Böhme, Jai Lal Lulla, Muhammad Auwal Abubakar, Christoph Treude, Sebastian Baltes

Research Collection School Of Computing and Information Systems

Agentic AI coding tools such as Claude Code and OpenAI Codex execute multi-step coding tasks with limited human oversight. To steer these tools, developers create repository-level configuration artifacts (e.g., Markdown files) for configuration mechanisms such as Context Files, Skills, Rules, and Hooks. There is no curated dataset yet that captures these configurations at scale. This dataset, collected from open-source GitHub repositories, fills that gap. We selected 40,585 actively maintained repositories through metadata filtering, classified them using GPT-5.2 to identify 36,710 as belonging to engineered software projects, and systematically detected configuration artifacts in these repositories. The dataset covers 4,738 repositories across …

Spatiotemporal Sycophancy: Negation-Based Gaslighting In Video Large Language Models, Ziyao Tang, Pengkun Jiao, Bin Zhu, Huiyan Qi, Jingjing Chen, Yu-Gang Jiang

Research Collection School Of Computing and Information Systems

Video Large Language Models (Vid-LLMs) have demonstrated remarkable performance in video understanding tasks, yet their robustness under conversational interaction remains largely underexplored. In this paper, we identify spatiotemporal sycophancy, a failure mode in which Vid-LLMs retract initially correct, visually grounded judgments and conform to misleading user feedback under negation-based gaslighting. Rather than merely changing their answers, the models often fabricate unsupported temporal or spatial explanations to justify incorrect revisions. To systematically investigate this phenomenon, we propose a negation-based gaslighting evaluation framework and introduce GasVideo-1000, a curated benchmark designed to probe spatiotemporal sycophancy with clear visual grounding and temporal reasoning requirements. …

Oscbench: Benchmarking Object State Change In Text-To-Video Generation, Xianjing Han, Bin Zhu, Shiqi Hu, Franklin Mingzhe Li, Patrick Carrington, Roger Zimmermann, Jingjing Chen

Research Collection School Of Computing and Information Systems

Text-to-video (T2V) generation models have made rapid progress in producing visually high-quality and temporally coherent videos. However, existing benchmarks primarily focus on perceptual quality, text–video alignment, or physical plausibility, leaving a critical aspect of action understanding largely unexplored: object state change (OSC) explicitly specified in the text prompt. OSC refers to the transformation of an object’s state induced by an action, such as peeling a potato or slicing a lemon. In this paper, we introduce OSCBench, a benchmark specifically designed to assess OSC performance in T2V models. OSCBench is constructed from instructional cooking data and systematically organizes action–object interactions into …

Tranx-Adapter: Bridging Artifacts And Semantics Within Mllms For Robust Ai-Generated Image Detection, Wenbin Wang, Yuge Huang, Jianqing Xu, Yue Yu, Jiangtao Yan, Shouhong Ding, Pan Zhou, Yong Luo

Research Collection School Of Computing and Information Systems

Rapid advances in AI-generated image (AIGI) technology enable highly realistic synthesis, threatening public information integrity and security. Recent studies have demonstrated that incorporating texture-level artifact features alongside semantic features into multimodal large language models (MLLMs) can enhance their AIGI detection capability. However, our preliminary analyses reveal that artifact features exhibit high intra-feature similarity, leading to an almost uniform attention map after the softmax operation. This phenomenon causes attention dilution, thereby hindering effective fusion between semantic and artifact features. To overcome this limitation, we propose a lightweight fusion adapter, TranX-Adapter, which integrates a Task-aware Optimal-Transport Fusion that leverages the Jensen-Shannon divergence …

Rendering Data Unlearnable By Exploiting Llm Alignment Mechanisms, Ruihan Zhang, Jun Sun

Research Collection School Of Computing and Information Systems

Large language models (LLMs) are increasingly trained on massive, heterogeneous text corpora, raising serious concerns about the unauthorised use of proprietary or personal data during model training. In this work, we address the problem of data protection against unwanted model learning in a realistic blackbox setting. We propose Disclaimer Injection, a novel data-level defence that renders text unlearnable to LLMs. Rather than relying on model-side controls or explicit data removal, our approach exploits the models’ own alignment mechanisms: injecting carefully designed alignment-triggers to prevent effective learning. Through layer-wise analysis, we find that finetuning on such protected data induces persistent activation …

Ontogenetic Variation In Palatal Morphology In Tufted And Untufted Capuchins, Valerie Contreras

Graduate Theses and Dissertations

ABSTRACT Craniofacial variation and its relation to diet have been widely examined, but with less attention exclusively to the palate. Previous studies have shown that groups consuming a more resistant diet exhibit morphological differences in the craniofacial skeleton, including a longer and wider palate when compared to groups consuming a less resistant diet; similar patterns have been observed among captive vs. wild populations. Here, I present a 3D morphometric analysis of palatal ontogeny in tufted (Sapajus spp.) and untufted (Cebus spp.) capuchins and compare captive vs. free-ranging semi-provisioned Sapajus. Tufted capuchins are known to consume more mechanically resistant foods compared …

Legal Terminology, Or: How Do Definitions Work For You?, Liliana Tymchenko

GHEA21 Publications

No abstract provided.

Dancing With Words: From My Clasp Notebooks, Dorota Ostrowska

GHEA21 Publications

No abstract provided.

Cultivating Critical Thinkers And Problem Solvers In Esl Classrooms Amid Political Unrest: Integrating Iwt Clasp Approaches, Reginald Louis

GHEA21 Publications

No abstract provided.

Colonization To Commodification To Codification: The Salimpour Method, Suhaila Salimpour

MFA in Dance Theses

This thesis examines how colonization and commodification have shaped the codification of belly dance though the Salimpour Method. Codification here refers to the process of naming, categorizing, and organizing of movement into a teachable system. While codification inevitably transforms tradition, it also serves as a method of preservation. As the inheritor of the Salimpour lineage, I trace how my mother, Jamila Salimpour, first codified belly dance and how I built on her work to create a comprehensive pedagogical system that situates belly dance within artistic and academic contexts. This study argues that the Salimpour Method is a strategy of preserving, …

#Metoo Vs. Anti-Dei, Nicole Porter

Faculty Publications

Nine years ago, the entire country was talking about the #MeToo movement, which can be credited with the unseating of hundreds of powerful men and changing the way we talk about and address sexual harassment in the workplace. In 2026, continued progress of the #MeToo movement might be threatened by a new movement-the Trump Administration's attack on Diversity, Equity, and Inclusion (DEI) initiatives. This article discusses the ways in which the #MeToo movement conflicts with the anti-DEI movement. Specifically, it addresses how employers will respond to pressure to dismantle DEI programs, offices, and training, and how those responses will affect …

Just Opt For Caesarean Section And Pick The Birthdate': A Qualitative Study Exploring Women's Delivery Preferences After Caesarean Section In Urban Tanzania, Daudi Gidion, Columba Mbekenga, Miriam Mgonja, Marleen Temmerman, Muzdalfat Abeid

Obstetrics and Gynaecology, East Africa

Objectives To explore perspectives, preferences and decision-making regarding mode of delivery after a previous caesarean section among women in urban Tanzania.

Design A descriptive qualitative study using semi-structured in-depth interviews. Data were analysed using thematic analysis.

Setting A private tertiary hospital in Dar es Salaam, Tanzania.

Participants 11 women with one previous caesarean section who delivered at the study site were interviewed on the third day after delivery. Participants were purposively sampled to include both modes of delivery: five women underwent repeat caesarean section and six attempted a trial of labour after caesarean (TOLAC).

Results Thematic analysis identified four key …

Health Beliefs And Lung Cancer Screening Uptake Among Eligible Asian Americans In Philadelphia, Pennsylvania, United States, Julia G. Katcher, Kuang-Yi Wen, Christine S. Shusted, Jessica Liang, Kyunghee Koh, Hee-Soon Juon, Julie Barta

Department of Medical Oncology Faculty Papers

OBJECTIVE: Asian Americans experience disproportionately late-stage lung cancer diagnoses and lower lung cancer screening (LCS) rates than other racial groups. This study evaluated the association between the

METHODS: This pilot randomized controlled trial, conducted April 2023-January 2024 within a Philadelphia health system, tested a culturally and linguistically tailored intervention led by community health workers. LCS completion and LCS-Health Belief Scale (LCS-HBS) responses were compared between control and intervention participants (

RESULTS: LCS completion was higher in the intervention group (45%) than in the control (31.3%). Control participants who completed LCS reported higher perceived lung cancer risk, greater perceived benefits, and …

Examining Data Richness In Undergraduate Students’ Reflections: A Linguistic Analysis Of Three Data Collection Methods, Kirsten A. Davis, Anne Wrobetz

School of Engineering Education Faculty Publications

Researchers select data collection methods for a variety of reasons. In our research on experiential education programs, we perceived that when we collected students’ reflections using different methods, the data we received was different. However, we found few existing studies to help us characterize what was different across our data sets to inform our decisions about which methods to use. Thus, the purpose of our study was to compare data richness across three data sets from previous studies of engineering students studying abroad, each collected using a different method (written diaries, interviews, video diaries). We used linguistic inquiry as a …

“Heavenly Fire Of The Godhead”: The Activity Of The Holy Spirit In The Macarian Writings, Daniel Wade Edwards

Dissertations (1934 -)

This dissertation addresses a lacuna within patristic scholarship on the doctrine of the Holy Spirit. Though the fourth century patristic writer known as “Macarius” addresses the theology of the Holy Spirit with unmatched depth and vibrancy, his pneumatological understanding has not yet been addressed adequately. Consequently, through a close literary analysis of the Spirit’s activity in the Macarian writings, this study proposes a threefold framework within which Macarius’ vision of the Spirit can be clearly perceived. In the first, the economic frame, the work of the Spirit within Old Testament Israel and the new covenant Church is compared and contrasted. …

Colonial Futures: Ngaio Marsh, War, And The End Of Empire, Heather M. Robinson

Publications and Research

Between the years of 1941 and 1945, Ngaio Marsh published three New Zealand-focused whodunnits; all three of these stories explore the impacts of the Second World War to some degree. What is curious is not that Marsh wrote about the war, but her sustained focus on New Zealand and New Zealanders, given that Marsh’s fiction usually took place in London or its surrounding counties. It marks the impact on Marsh of the global conflict that was changing the world, and shaking the foundations of colonialism and modernity as they had been known before 1939. In this essay, I explore how …

Automated Scoring Of Prosody In Oral Reading Fluency: Integrating Acoustic And Sentence/Chunk-Level Silence Features With Machine Learning, Kuo Wang

Education Policy and Leadership Theses and Dissertations

Prosody, the ability to read with appropriate variation in pitch, duration, pausing, and stress, is an important (and challenging to measure) dimension of oral reading fluency (ORF) and comprehension. ORF prosody is currently measured through human raters and rubric-based scoring systems (e.g., the National Assessment of Educational Progress (NAEP) or the Multidimensional Fluency Scale (MDFS scales). The rating process is labor and resource-intensive and introduces rater bias. Recent work has shown acoustic features to predict prosody ratings; however, their models had low generalization performance on cross-domain data (unseen passages and student samples).

The present study investigated the automated assessment of …

In The Shadow Of The Steeple: A Theory Of Religious Hypocrisy, Justin Arthur Rice

Graduate Theses and Dissertations

In this dissertation, I explore how Christian doctrine and practice respond to the pressures placed upon them by the troubling reality of religious hypocrisy (RH). Assuming a teleological notion of social institutions, I argue that RH is best understood in relation to the Church as an institution, and that RH occurs when religious agents undermine or subvert their appointed role within the Church. RH consists in the conscious or unconscious charade of spiritual or moral virtue undertaken in one’s capacity as a member of a religious institution, grounded in a fundamental disparity between one’s thoughts, words, desires, motivations, or deeds, …

Physics-Informed Machine Learning For Predictive Digital Twins In Greenhouses, Hoang Kim Tran

Graduate Theses and Dissertations

Greenhouses are widely used to create controlled environments for crop production, where indoor climate variables such as air temperature, relative humidity, CO2 concentration, radiation, and soil or substrate conditions directly affect plant growth, yield, energy consumption, and resource use. However, greenhouse climates are highly dynamic because they are influenced by complex interactions among outdoor weather, greenhouse structure, crop transpiration, heating systems, ventilation, screens, fans, irrigation, and other actuators. Building digital twins for greenhouses provides a powerful way to represent, monitor, and predict these climate dynamics. A predictive digital twin can simulate future indoor climate states based on current greenhouse conditions, …

Localizing And Repairing Backdoor-Sensitive Layers In Pre-Trained Language Models For Secure Fine-Tuning, Sunanda Das

Graduate Theses and Dissertations

Task-agnostic backdoor attacks can contaminate pre-trained language models (PLMs) in a way that survives downstream adaptation, even under full fine-tuning, making it difficult for practitioners to trust third-party checkpoints. Existing defenses often rely on privileged assumptions (e.g., access to poisoned data or trigger/target knowledge), thereby limiting their applicability in realistic settings. We present DiSec (Disentanglement of potentially adversarial weights for Secure fine-tuning), a robust and label-efficient purification framework that uses only clean auxiliary text and does not rely on downstream supervision or attack signatures. DiSec elicits model-internal signals from this clean data to separate suspicious parameter components that are inconsistent …

There Is No Free Benchmark: An Institutional View Of Legal Ai Benchmarking, Neel Guha, Andy K. Zhang, Christine Tsang, Christopher D. Manning, Julian Nyarko, Daniel E. Ho

Faculty Scholarship

Despite substantial excitement around the use of AI in law, little information exists on the performance and associated risks of the domain’s widely marketed tools. Recent work, for instance, has demonstrated the significant potential for “hallucinations” — wherein models make up facts, law, and precedent — leading Chief Justice Roberts to spotlight this risk in his annual report on the judiciary. We argue that there is a need for public AI benchmarking in law. First, relative to other AI application domains, the legal AI ecosystem lacks legibility — there is little information about the design and performance of many commercial …

From Learner To Leader: Exploring Principals' Perspectives On Professional Learning, Development, And Mentoring, Eric William Kausch

Graduate Theses and Dissertations

This purpose of this qualitative study was to examine veteran secondary school principals’ engagement in professional learning, development, and mentoring for the improvement of their leadership skills. While previous literature provides insight into administrative preparation programs, mentoring of novice leaders, and succession planning for those advancing their careers, there are fewer studies that discuss enhancing experienced principals’ leadership development. The study was guided by three research questions: (1) How do principals describe their experiences participating in professional learning and development? (2) What types of leadership experiences do principals engage in to improve their leadership skills? and (3) How does engaging …

Drawing Conclusions: Autonomy And Mental Maps In Denmark And The Faroe Islands, Erik Steven Breidinger

Graduate Theses and Dissertations

This dissertation develops and tests a comparative heuristic device called optimal autonomy to explain when and how autonomous territories, or simply autonomies, become durable components of the global system of sovereign states. Drawing on a mixed‑methods case study of Denmark and the Faroe Islands, the study operationalizes optimal autonomy as a heuristic device using three interacting metrics: territorial delineation, national distinction, and state dependence. Using semi‑structured interviews with public officials and a mental‑mapping experiment with university students, this research shows that optimal autonomies are not made from single institutional designs but rather a dynamic equilibrium between maximal territorial delineation, medial …

A Dual Analysis Of Generic Stereotypes, Shaoneng Xu

Graduate Theses and Dissertations

In my thesis, I provide a dual-analysis on generic stereotypes, including a semantic analysis that focuses on the truth conditions of generic sentences as well as a pragmatic analysis that focuses on the normative force of generic stereotypes. The thesis contains two parts: In the first chapter, I make a semantic analysis of generic stereotypes based on Lemeire’s (2020) paper. I unpack the general concept of generic sentences and their truth-conditions. I also discuss Lemeire’s rebuttal strategies to offensive generic stereotypes, and add to it by indicating that generics can also carry normative forces. In the second chapter, I make …

Advancing Social Media Analytics And Personalized Generation Via Transfer Learning, Discourse-Aware Modeling, And Collaborative Modeling, Gibson Nkhata

Graduate Theses and Dissertations

Social media platforms have become central to information exchange, shaping public opinion across social, political, and economic domains. However, the massive volume of user-generated content, combined with its informal, nuanced, and often noisy nature, presents significant challenges for automated analysis and generation. Tasks such as stance detection, rumor verification, and personalized content generation are further complicated by sarcasm, evolving discourse structures, and diverse user preferences. Addressing these challenges requires models that can effectively leverage linguistic nuance, conversational dynamics, and collaborative user signals. Transfer learning has emerged as a powerful paradigm for improving performance in low-resource and complex language understanding tasks. …

Broadband Diffusion And Digital Preparedness In Tanzania: Harnessing The Role Of Technology In Formal Education, George Matto, John J. Iwata

Journal of Global Education and Research

Broadband, as a high-speed Internet connectivity, is an essential enabler for digital preparedness. However, little is known, in the case of Tanzania, about the nexus between broadband diffusion and digital preparedness, particularly in the facilitation of the teaching and learning process. This study, therefore, aimed at investigating the nexus between broadband diffusion and digital preparedness in Tanzania by focusing on the country’s readiness to leverage the same for formal education. The study employed a systematic review, guided by the PRISMA framework, to obtain the empirical evidence from existing literature. Thematic analysis was used to identify themes, analyze, and interpret qualitative …

Sjsu Erfa News, Summer 2026, San Jose State University, Emeritus And Retired Faculty Association

Emeritus and Retired Faculty Association (ERFA) Newsletter

Volume 39, No 39

Medadl: High-Throughput Information Extraction Of Functional Status From Electronic Health Records To Advance Frailty Assessment In Older Adults, Sunyang Fu, Zhiyi Yue, Jennyly T Nguyen, Huipeng Liu, Jaerong Ahn, Vanessa Ramirez, Jude Des Bordes, Hongfang Liu, Min Ji Kwak, Nahid J Rianon

Faculty, Staff and Student Publications

Background: Functional status is essential for assessing frailty and planning care in older adults but is often under-documented in the structured fields of electronic health records. Manual chart review can capture functional status information but is labor-intensive and time-consuming. In this study, we developed and validated a scalable natural language processing (NLP) model to extract functional status information from unstructured electronic health record (EHR) notes.

Methods: This retrospective cohort study included 110 older adults seen at a geriatric osteoporosis clinic. A large language model-augmented symbolic NLP pipeline (MedADL) was developed to extract activities of daily living (ADL) and instrumental activities …

Implementation Brief 2 Actionable Foundational Practices Tool, Center On English Language Proficiency

Resources

No abstract provided.