AUSpace

AUSpace is Athabasca University’s institutional repository - an open access digital service that collects, preserves, and provides access to the intellectual, educational, and institutional outputs of the university community.

For more information, please visit our About page.

Communities in AUSpace

Select a community to browse its collections.

Now showing 1 - 9 of 9

Recent Submissions

Access status: Open Access ,
PARAMETER-EFFICIENT WORD SENSE DISAMBIGUATION THROUGH LOW- RANK AND DECOMPOSED ADAPTATION
(2026-09-09) Manikandan, Vijayalakshmi; Wen, Dunwei; Dewan, Ali; Valderrama, Camilo
Word Sense Disambiguation (WSD) remains a fundamental challenge in natural language processing (NLP), necessitating models that can resolve ambiguity across both general and specialized domains. While fine-tuning large pre-trained language models has yielded good performance gains, these approaches are often parameter-inefficient, requiring the update of millions of model parameters for task-specific adaptation. This research proposes a unified, parameter-efficient framework for WSD that addresses these limitations through innovative architectural modules and domain-specific optimization strategies. For general- domain WSD, we introduce an approach utilizing Low-Rank Adaptation (LoRA) modules coupled with a novel Part-of-Speech (POS) retrieval pipeline. This pipeline aligns target word contexts with WordNet lexical categories to construct optimized sentence-gloss pairs, enabling the model to outperform state-of-the-art benchmarks on Senseval and SemEval datasets using only 0.5% of the original model parameters. Extending this framework to the biomedical domain, where abbreviation ambiguity and context-dependent terminology present unique hurdles, we propose a multi-task learning architecture that jointly optimizes classification and ranking objectives. To enhance discriminative learning in this high-density semantic environment, we implement a category-aware balanced sampling strategy and structured negative sampling, which incorporates both hard and semantically similar negatives. By employing Weight- Decomposed Low-Rank Adaptation (DoRA), the model achieves superior adaptation while updating a mere 0.29% of total parameters. Extensive empirical evaluations on the MeDAL and MSH datasets demonstrate state-of-the-art performance, achieving a 92.9% macro F1 score and 98.1% accuracy, respectively. Systematic ablation studies further validate that the integration of multi-task structured objectives, balanced data sampling, and parameter-efficient adaptation yields a robust, scalable framework. By harmonizing these elements, this research establishes a new benchmark for WSD, proving that high- precision semantic disambiguation can be achieved with minimal computational overhead across diverse linguistic landscapes. This dual-domain approach provides a validated pathway for future advancements in specialized medical text applications and general- purpose natural language understanding.
Access status: Open Access ,
Distinguishing Transmisogyny from Transphobia in Counselling Contexts
(2026-09-07) Hugessen, Han; Doyle, Emily; Nylund, David; Edwards, Margaret
There is a critical yet undertheorized distinction in gendered oppression: the difference between transphobia as a broad descriptor of discrimination against trans people and transmisogyny as a specific, structurally patterned form of violence directed at transfeminine lives. Counselling literature and healthcare guidelines continue to subsume it under the umbrella of transphobia, leaving its unique mechanisms, impacts, and clinical relevance largely unarticulated. This project makes visible the counselling practices that flatten trans experience, omit distinctions between transmasculine and transfeminine clients, and rely on frameworks that implicitly centre those who benefit from patriarchal privilege. This work analyses the awareness and inclusion of transmisogyny in clinical guidelines and research, finding limited relevant frameworks for counsellors to adequately address the unique positioning of their transfemme clients.
Access status: Open Access ,
Data Management Plan: Dr. Katie MacDonald
(Athabasca University Library and Scholarly Resources, 2026-08-25) MacDonald, Katie; Wang, Dandi; Stobbs, Robyn
This Data Management Plan example is intended as a tool for research data management, and references one of Dr. Katie MacDonald's research with housing organizations to understand how the housing sector is addressing housing needs and how they can strengthen the public housing sector in Canada. This DMP example was created for the <a href="https://pressbooks.openeducationalberta.ca/rdmtoolkit/">Research Data Management and the Cloud: A Toolkit for Researchers</a> Pressbook.
Access status: Open Access ,
File-Naming and Organization Worksheet
(Athabasca University Library and Scholarly Resources, 2026-08-25) Stobbs, Robyn
This is a worksheet designed to assist researchers with file-naming and organization of folders as a part of their data documentation practices. It includes recommendations for developing systematic file-naming conventions. This worksheet was created for the <a href="https://pressbooks.openeducationalberta.ca/rdmtoolkit/">Research Data Management and the Cloud: A Toolkit for Researchers</a> Pressbook.
Access status: Open Access ,
Data Management Plan: Dr. Srijak Bhatnagar
(Athabasca University Library and Scholarly Resources, 2026-08-25) Bhatanagar, Srijak; Wang, Dandi; Stobbs, Robyn
This Data Management Plan example is intended as a tool for research data management, and references one of Dr. Srijak Bhatnagar's research to study environmental contamination in terrestrial and aquatic ecosystems. This DMP example was created for the <a href="https://pressbooks.openeducationalberta.ca/rdmtoolkit/">Research Data Management and the Cloud: A Toolkit for Researchers</a> Pressbook.