About

I’m Professor of Natural Language Processing at the University of Cambridge, co-Director of the Language Technology Lab (LTL), and Chief Scientist of Trismik. My work focuses on language, safety, and evaluation for AI systems — with an emphasis on uncertainty, calibration, and human-centred assessment of model behaviour.

Profiles: LinkedIn · Google Scholar · ORCID · LTL Lab

Reflections

Occasional essays exploring humanistic perspectives on AI — what it means for knowledge, creativity, trust, and the future of scholarship.

• (Coming soon) Uncertainty is a feature, not a bug
• (Coming soon) Uncertainty as a feature, not a bug

Research

Short, readable notes on recent work from our group. Selected 2025–2023 papers:

2025

• UNCLE: Benchmarking Uncertainty Expressions in Long-Form Generation
  R. Yang, C. Zhang, Z. Zhang, X. Huang, D. Yu, N. Collier, D. Yang — EMNLP 2025. Preprint: arXiv:2505.16922
• SimBench: Benchmarking the Ability of LLMs to Simulate Human Behaviours
  T. Hu, J. Baumann, L. Lupo, N. Collier, D. Hovy, P. Röttger — Preprint: arXiv:2510.17516
• Conformity in Large Language Models
  X. Zhu, C. Zhang, T. Stafford, N. Collier, A. Vlachos — ACL 2025.
• LOGU / Long-form Generation with Uncertainty Expressions
  R. Yang, C. Zhang, Z. Zhang, X. Huang, S. Yang, N. Collier, D. Yu, D. Yang — ACL 2025. Preprint: arXiv:2505.16922
• 500xCompressor: Generalized Prompt Compression for LLMs
  Z. Li, Y. Su, N. Collier — ACL 2025.
• iNews: An Individualized News Affective Response Dataset
  T. Hu, K. Garimella, N. Collier — 2024/2025 dataset releases.
• When Personalization Meets Reality: A Multi-Faceted Analysis of Personalized Preference Learning
  Y. R. Dong, T. Hu, Y. Liu, A. Üstün, N. Collier — Preprint: arXiv:2502.19158
• Nature Computational Science: Generative Language Models Exhibit Social Identity Biases
  T. Hu, Y. Kyrychenko, S. Rathje, N. Collier, S. van der Linden, J. Roozenbeek — Nat. Comput. Sci. 5(1):65–75, 2025.

2024

• LUQ: Long-Text Uncertainty Quantification — Preprint: arXiv:2403.20279
• Quantifying the Persona Effect in LLM Simulations — Preprint: arXiv:2402.10811

People

PhD researchers in the Language Technology Lab (Cambridge):

• Meiru Zhang — PhD student, Gates Cambridge Scholar, D4
• Yinhong Liu — PhD student, D4
• Tiancheng Hu — PhD student, Gates Cambridge Scholar, D3
• Chang Shu — PhD student, D3
• Caiqi Zhang — PhD student, D3
• Yijiang River Dong — PhD student, D2
• Zongqian Li — PhD student, D2
• Sanhanat Sivapiromrat — PhD student, D2
• Auss Abbood — PhD student, D2
• Zack Hui — PhD student, D1

Lab website: ltl.mmll.cam.ac.uk

Bio

Nigel H. Collier (ORCID 0000-0002-7230-4164) is Professor of Natural Language Processing at the University of Cambridge and co-Director of the Language Technology Lab. Prior roles include Associate Professor at the National Institute of Informatics, Tokyo. He leads collaborative research on reliability, uncertainty, and alignment in large language models.