Elamur Research⌕
☰
home / authors / Catherine Arnett

Catherine Arnett

Catherine Arnett has 4 papers in our library, in 2024–2026, most of them on Language models & pretraining, Multilingual & cross-lingual NLP, Low-resource languages; often with Benjamin Bergen, Akshat Gupta, Atnafu Lambebo Tonja.

4 papers in our library · 2024–2026

OpenAlex · h-index 3 · 41 works · 46 citations OpenAlex, 2026-10-14

📄 4 works in our library · 🔗 Cited 119 times in our library · 1 more in our library without a page yet · ⭐ Paper Stars ranking: #6267

The profile is the library's match of the name; two people with one name are not always told apart.

Elsewhere: Google Scholar · Semantic Scholar · dblp · ACL Anthology

 Affiliations

🎓University of California, BerkeleyBerkeley · United States2026 · 3 paperscurrent🎓Massachusetts Institute of TechnologyCambridge · United States2026 · 2 paperscurrent🎓University of CambridgeCambridge · United Kingdom2026 · 2 paperscurrent🏢IBMArmonk · United States2026 · 1 papercurrent🎓University of California San DiegoSan Diego · United States2023–2024 · 8 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine.

 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Artificial Intelligence · 4 worksArts and Humanities · 4 worksComputer Science · 4 worksLanguage and Linguistics · 4 works

🧭 Topics 4Research subjects and languages the work is about.Language models & pretraining · 2 worksMultilingual & cross-lingual NLP · 2 worksLow-resource languages · 1 workMorphology · 1 work

 Works by year

2020s

20262 papers20251 paper20241 paper

 Papers

The most cited first (OpenAlex and our library).

  1. When Is Multilinguality a Curse? Language Modeling for 250 High- and Low-Resource Languages (2024 · 87 citations)
    Multilingual language models are often used to support low-resource languages, but there is little direct evidence about how sharing a model with other languages affects modeling quality in any single language.
  2. Why do language models perform worse for morphologically complex languages? (2025 · 53 citations)
    Language models tend to perform better on fusional languages such as English than on agglutinative ones such as Turkish.
  3. CommonLID: Re-evaluating State-of-the-Art Language Identification Performance on Web Data (2026 · 3 citations)
    Language identification (LID) is often treated as a solved task. In practice, it remains unreliable for many languages on noisy, mixed web text, which is the main source of multilingual pre-training corpora. Most existing benchmarks, such as FLORES, UDHR and Bible data, are clean, translated or religious, so they…
  4. Weight Tying Biases Token Embeddings Towards the Output Space (2026)
    Many language models share one matrix between the input embedding and the output (unembedding) layer, but it has been unclear what this sharing does to the learned representations.

All works

5 works of theirs in our library, the most cited first; 4 with a page here, the others link their original until we summarise them.

  1. When Is Multilinguality a Curse? Language Modeling for 250 High- and Low-Resource LanguagesConference on Empirical Methods in Natural Language Processing (EMNLP)Tyler A. Chang, Catherine Arnett, Zhuowen Tu, et al.2024 · 87 citations
  2. Why do language models perform worse for morphologically complex languages?International Conference on Computational Linguistics (COLING)Catherine Arnett, Benjamin Bergen2025 · 53 citations
  3. CommonLID: Re-evaluating State-of-the-Art Language Identification Performance on Web DataAnnual Meeting of the Association for Computational Linguistics (ACL)Pedro Ortiz Suarez, Laurie Burchell, Catherine Arnett, et al.2026 · 3 citations
  4. Soohak: A Mathematician-Curated Benchmark for Evaluating Research-level Math Capabilities of LLMs ↗arXiv:2605.09063 · no summary yetGuijin Son et al.2026 · 1 citation
  5. Weight Tying Biases Token Embeddings Towards the Output SpaceFindings of the Association for Computational LinguisticsAntonio Lopardo, Avyukth Harish, Catherine Arnett, et al.2026

 Co-authors here

11 co-authors with a page, the most papers together first.

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

 Cited by

Papers of our library cite their works 119 times (of the 149,439 papers whose bibliographies we read); 29 citing papers with a page here →.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/authors/catherine-arnett/