About Me
I am an NLP and computational social science researcher studying how word meanings change across time and across scientific, media, and public language. My work combines historical text corpora, contextual embeddings, lexical resources, statistical modeling, and large language models to study semantic and conceptual change in mental-health language, media discourse, and socially contested language. I bring social science and psychology training to NLP questions about meaning, concepts, and cultural change, especially where language reflects shifting social and psychological categories. I led the development of SIBling and LSC-Eval, frameworks for modeling and evaluating lexical semantic change. More broadly, I am interested in language change and variation, shifting mental health concepts, human–LLM evaluation, and computational approaches to culturally significant questions. My work has been supported by the Australian Government Research Training Program Scholarship, Concept Creep research funding, and Change is Key! research support.
- Natural Language Processing
- Computational Social Science
- Semantic Change
- Social Psychology
- Concept Creep & Mental Health
PhD, Natural Language Processing & Social Psychology
University of Melbourne (2023-26)
Graduate Diploma in Psychology (Advanced) with Honours
University of Melbourne
I study how meanings shift as concepts move across scientific, media, and general-language contexts, especially when language carries social, cultural, or psychological significance. My work asks how mental health concepts and socially contested terms extend into new contexts, change in severity or emotional force, and acquire new evaluative associations. To examine these questions, I combine NLP methods with theory and concepts from social psychology, computational social science, and corpus linguistics, using historical text corpora, contextual embeddings, lexical resources, large language models, and statistical modeling.
Research streams
My work asks how meanings and concepts change as they move across scientific, media, and public contexts, especially where language reflects shifting boundaries between normality and pathology, harm and hardship, clinical and everyday meaning, or human and dehumanized social categories. It clusters around three overlapping streams:
Semantic and conceptual change
Developing methods for tracing how meanings shift across time, from multidimensional models of lexical semantic change to synthetic benchmarks, diachronic word-sense tracking, and recent work on how humans and language models represent meaning.
Key work: SIBling, LSC-Eval, Diachronic WSD, SenseRel.Mental health concepts and category boundaries
Applying these methods to mental health concepts and terminology, examining how meanings and representations shift across psychology, news media, books, and general American English, interpreting findings through the lens of psychological theory.
Selected work: Historical semantic shifts in trauma, schizophrenia, generic and emotion-related mental-health terminology, and collaborative projects on ADHD, anxiety and depression.Socially contested language
Studying how language encodes negative evaluation, dehumanization, identity, stigma, and contested social meanings in public discourse.
Selected collaborative work: Dehumanization of women in incel discourse, mental health stigma detection in online communication, identity/person-first language for mental health conditions, and laypeople’s understandings of the common good.
ACL 2026: Joint-first author on SenseRel, a sense-level benchmark for denotational and connotational meaning relations, developed through my Change is Key! research internship.
Findings of ACL 2025: Lead author of LSC-Eval, a framework for evaluating methods for detecting dimensions of lexical semantic change using LLM-generated synthetic data.
ACL 2024: Lead author of SIBling, a multidimensional framework for modeling lexical semantic change with social science applications.
ICWSM 2026: Lead author of a computational social science paper on dehumanization of women and men in incel discourse.
Awards and research support: Australian Government Research Training Program Scholarship; selected for Change is Key! research support for an international research internship and collaboration; supported by Concept Creep research funding for conference travel and research dissemination.
Competitive team funding: Co-recipient of AUD $15,000 Hallmark Research Initiative seed funding for a 7-member team project on ‘Automatic evaluation of mental health stigma in online communication’, led by Yulia Otmakhova.
Research leadership and service: Australian English Language Co-Lead with Christine de Kock for the BLEnD SemEval-2026 Shared Task; Program Chair/PC for the LChange'26 Workshop, co-located with EACL 2026.
ACL Best Resource Paper Award: Contributor to BRIGHTER: BRIdging the Gap in Human-Annotated Textual Emotion Recognition Datasets for 28 Languages, awarded Best Resource Paper at ACL 2025; contributed Romanian-language gold-standard annotation (team lead: Daniela Teodorescu).
Invited talks: Presented PhD work on modeling semantic change in mental-health concepts at international research groups: Utrecht University, the University of Gothenburg, and the National Research Council Canada.
My work is motivated by substantive questions about language, meaning, mental health, and the shifting boundaries of social and psychological categories. I use computational methods and careful measurement to better understand how these categories change, and to make clearer, responsible claims about socially important questions. Theory only becomes useful empirically when we can make clear, interpretable, and well-grounded claims about what we are measuring.