🧑‍🔬 Willem Zuidema

🏛️ Affiliations

🎓University of AmsterdamAmsterdam · Netherlands2006–2026 · 95 papersEmployment · Associate Professor, ILLCcurrent🎓Vrije Universiteit BrusselBrussels · Belgium1999–2002 · 13 papers🎓University of EdinburghEdinburgh · United Kingdom2002–2003 · 7 papers🎓Leiden UniversityLeiden · Netherlands2008–2010 · 4 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 1 more institution on single bylines are not listed.

🧭 Fields, topics and concepts

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

📅 Works by year

2020s

📄 Papers

The most cited first (OpenAlex and our library).

1. Quantifying Attention Flow in Transformers (2020 · 1,064 citations)
   Raw self-attention weights are often used to explain Transformer decisions. In higher layers, however, every embedding already mixes information from many tokens, so these weights say little about which input tokens mattered. Abnar and Zuidema treat the attention across layers as a directed acyclic graph: nodes are…
2. Structural Persistence in Language Models: Priming as a Window into Abstract Language Representations (2022 · 74 citations)
   Do neural language models represent sentence structure abstractly, apart from the specific words used? The authors test this with structural priming, a method from psycholinguistics: after a model reads a sentence with a given construction, does it assign higher probability to a follow-up sentence built the same way?…
3. Identifying and Adapting Transformer-Components Responsible for Gender Bias in an English Language Model (2023 · 19 citations)
   The paper asks whether gender bias in a language model can be traced to specific internal parts and then reduced by retraining only those parts.
4. The Birth of Bias: A case study on the evolution of gender bias in an English language model (2022 · 19 citations)
   Most work on gender bias in language models detects or removes it after training. This paper asks a prior question: how does a model become biased while it learns? The authors chose a small, fully transparent setup, an LSTM language model trained on an English Wikipedia corpus. They had complete access to the…

All works

7 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. Quantifying Attention Flow in TransformersAnnual Meeting of the Association for Computational Linguistics (ACL)Samira Abnar, Willem Zuidema2020 · 1,064 citations
2. Structural Persistence in Language Models: Priming as a Window into Abstract Language RepresentationsTransactions of the Association for Computational Linguistics (TACL)Arabella Sinclair, Jaap Jumelet, Willem Zuidema, et al.2022 · 74 citations
3. BLiMP-NL: A Corpus of Dutch Minimal Pairs and Acceptability Judgments for Language Model Evaluation ↗Computational Linguistics · no summary yetMichelle Suijkerbuijk et al.2025 · 22 citations
4. Identifying and Adapting Transformer-Components Responsible for Gender Bias in an English Language ModelBlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLPAbhijith Chintam, Rahel Beloch, Willem Zuidema, et al.2023 · 19 citations
5. The Birth of Bias: A case study on the evolution of gender bias in an English language modelWorkshop on Gender Bias in Natural Language ProcessingOskar Van Der Wal, Jaap Jumelet, Katrin Schulz, et al.2022 · 19 citations
6. Learning from errors: Using vector-based compositional semantics for parse reranking ↗Proceedings of the Workshop on Continuous Vector Space Models and their Compositionality · no summary yetPhong Le, Willem Zuidema, Remko Scha2013 · 8 citations
7. What do self-supervised speech models know about Dutch? Analyzing advantages of language-specific pre-training ↗arXiv:2506.00981 · Proc. INTERSPEECH 2025 · no summary yetMarianne de Heer Kloots et al.2025 · 2 citations

🤝 Co-authors here

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

• Jaap Jumelet ×2
• Oskar van der Wal ×2
• Arabella Sinclair
• Michael Hanna
• Raquel Fernández
• Samira Abnar

Venues

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

• Annual Meeting of the Association for Computational Linguistics (ACL): ACL 2020
• BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP: BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP 2023
• Workshop on Gender Bias in Natural Language Processing: Workshop on Gender Bias in Natural Language Processing 2022
• Transactions of the Association for Computational Linguistics (TACL): Transactions of the Association for Computational Linguistics (TACL) 2022
• Computational Linguistics
• Proceedings of the Workshop on Continuous Vector Space Models and their Compositionality

🔗 Cited by

Papers of our library cite their works 220 times (of the 149,439 papers whose bibliographies we read); 32 citing papers with a page here →.

Updated 30 October 2026

↑ back to top

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