MorphPiece : A Linguistic Tokenizer for Large Language Models

Tokenization is a critical part of modern NLP pipelines. However, contemporary tokenizers for Large Language Models are based on statistical analysis of text corpora, without much consideration to the linguistic features. I propose a linguistically motivated tokenization scheme, MorphPiece, which is based partly on morphological segmentation of the underlying text. A GPT-style causal language model trained on this tokenizer (called MorphGPT) shows comparable or superior performance on a variety of supervised and unsupervised NLP tasks, compared to the OpenAI GPT-2 …

analysis cs.cl features language language models large language large language models modern nlp part pipelines segmentation statistical text tokenization