Revisiting subword tokenization: A case study on affixal negation in large language models

arXiv:2404.02421v1 Announce Type: new
Abstract: In this work, we measure the impact of affixal negation on modern English large language models (LLMs). In affixal negation, the negated meaning is expressed through a negative morpheme, which is potentially challenging for LLMs as their tokenizers are often not morphologically plausible. We conduct extensive experiments using LLMs with different subword tokenization methods, which lead to several insights on the interaction between tokenization performance and negation sensitivity. Despite some interesting mismatches between tokenization accuracy …

abstract arxiv case case study cs.cl english impact language language models large language large language models llms meaning modern negative study through tokenization type work