Repetition Neurons: How Do Language Models Produce Repetitions?
Hiraoka, Tatsuya Inui, Kentaro
Hiraoka, Tatsuya
Inui, Kentaro
Author
Supervisor
Department
Natural Language Processing
Embargo End Date
Type
Conference proceeding
Date
License
Language
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
This paper introduces repetition neurons, regarded as “skill neurons” responsible for the repetition problem in text generation tasks. These neurons are progressively activated more strongly as repetition continues, indicating that they perceive repetition as a task to copy the previous context repeatedly, similar to in-context learning. We identify these repetition neurons by comparing activation values before and after the onset of repetition in texts generated by recent pre-trained language models. We analyze the repetition neurons in three English and one Japanese pre-trained language models and observe similar patterns across them.
Citation
Source
Proceedings of the 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies: Long Papers, NAACL-HLT 2025
Conference
Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers)
Keywords
Subjects
Source
Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers)
Publisher
Association for Computational Linguistics (ACL)