Nina Skovgaard Schneidermann

PhD Fellow | Computational Linguist | Accidental Accessibility Architect

Summary

PhD Fellow in Computational Linguistics specializing in the computational modeling of non-literal language for Danish, including hyperbole and metaphor detection.

Research and consultancy work focuses on:

  • NLP for low-resource languages
  • Explainable AI (XAI)
  • LLM evaluation
  • linguistic datasets
  • accessibility and inclusive design

Combines computational linguistics expertise with extensive experience navigating and auditing digital accessibility systems as a blind researcher and consultant.

Education

PhD in Computational Linguistics

University of Copenhagen — Center for Language Technology (CST)
2021 – Present

Research Focus

  • Figurative language modeling for Danish
  • Explainable AI approaches for NLP
  • Hyperbole and metaphor interpretation

Visiting Researcher Vrije Universiteit Amsterdam — CLTL Lab
Feb 2023 – Aug 2023

Collaborated with Prof. Piek Vossen on:

  • metaphor analysis
  • frame semantics
  • computational linguistics research

MSc in IT and Cognition

University of Copenhagen
2016 – 2019

BA in Linguistics

University of Copenhagen
2013 – 2016

Exchange Program

University of California, Davis (UC Davis)
March 2016 – August 2016

Experience

Academic Teaching & Lecturing

Co-Lecturer — Information Retrieval & Programming (BA Level)

University of Copenhagen
Feb 2022 – July 2022

  • Co-taught semester course in:
    • programming fundamentals
    • information retrieval
    • computational methods
  • Delivered guest lecture on:
    • sentiment analysis
    • text mining

Guest Lecturer — Digital Design (MA Level)

IT University of Copenhagen (ITU)
April 2026

  • Delivered invited lecture on:
    • digital accessibility
    • inclusive design
    • accessibility workflows

Research & Consultancy

PhD Researcher

University of Copenhagen
2021 – Present

  • Architecting foundational datasets for figurative language in Danish
  • Building evaluation benchmarks for non-literal language understanding
  • Evaluating multi-task learning architectures for low-resource NLP tasks

Linguistic Consultant & Data Architect

Contract / Freelance
2017 – Present

  • Curating and annotating Nordic linguistic resources
  • Developing datasets for:
    • speech recognition
    • sentiment analysis
    • Danish word embeddings

Accessibility & AI Consultant

Freelance / Project-Based
2018 – Present

  • Evaluating AI-generated image captions for accessibility
  • Conducting WCAG accessibility audits
  • Assessing cross-cultural accessibility systems
  • Developing NLP workflows compatible with:
    • JAWS
    • NVDA
    • VoiceOver

Publications

2026

DAMETA: An LLM Benchmark for Danish Metaphor Interpretation with Systematically Varied Distractors
Schneidermann, N. S., Nimb, S., Norman, N. C. H., Olsen, S., & Pedersen, B. S. (2026). DAMETA: An LLM Benchmark for Danish Metaphor Interpretation with Systematically Varied Distractors. I Proceedings of Language Resources and Evaluation Conference [https://lrec.elra.info/lrec2026-main-930]

2025

Lexical-Semantic Resources as a Culture-Aware Basis for Benchmarking and Evaluation of LLMs
Norman, N., Nimb, S., Olsen, S., Schneidermann, N., & Pedersen, B. S. (2025). Lexical-Semantic Resources as a Culture-Aware Basis for Benchmarking and Evaluation of LLMs. Electronic lexicography in the 21st century (eLex 2025), pp. 517–533.

2023

Probing for Hyperbole in Pre-Trained Language Models
Schneidermann, N., Hershcovich, D., & Pedersen, B. (2023). Probing for Hyperbole in Pre Trained Language Models. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics - - Student Research Workshop (ACL-SRW 2023), pp. 200–211.

2022

Evaluating a New Danish Sentiment Resource: The Danish Sentiment Lexicon (DSL)
Schneidermann, N., & Pedersen, B. (2022). Evaluating a New Danish Sentiment Resource: The Danish Sentiment Lexicon, DSL. Proceedings of the 2nd Workshop on Sentiment Analysis and Linguistic Linked Data (SALLD-2), pp. 19–24.

2020

Towards a Gold Standard for Evaluating Danish Word Embeddings
Schneidermann, N., Hvingelby, R., & Pedersen, B. (2020). Towards a Gold Standard for Evaluating Danish Word Embeddings. Proceedings of the 12th Language Resources and Evaluation Conference (LREC 2020), pp. 4754–4763.

Funding & Awards

  • Martin Levys Mindelegat
  • Knud Højgaards Fond
  • William Demand Fond
  • Lizzi & Mogens Staal Fond

Funding awarded for:

  • conference participation
  • academic travel
  • research support

Skills & Expertise

Machine Learning & NLP

Deep Learning

  • Transformers
    • BERT
    • RoBERTa
  • LLM Fine-Tuning
  • Prompt Engineering
  • LangChain
  • Ollama

Frameworks & Libraries

  • PyTorch
  • TensorFlow
  • Hugging Face Transformers
  • SpaCy
  • Scikit-learn

Research Areas

  • Multi-task learning
  • Transfer learning
  • Sentiment analysis
  • Explainable AI

Technical Skills

  • Python
  • SQL
  • Unix/Linux
  • Git
  • Docker

Accessibility Technologies

Expert user and evaluator of:

  • JAWS
  • NVDA

  • VoiceOver

Languages

  • Danish (Native)
  • English (Fluent)
  • German (Working proficiency)