paper-with-me

Papers

Multi-word Tokenization for Sequence Compression

2024-02-15 · Leonidas Gee, Leonardo Rigutini, Marco Ernandes, Andrea Zugarini

Large Language Models have proven highly successful at modelling a variety of tasks. However, this comes at a steep computational cost that hinders wider industrial uptake. In this paper, we present MWT: a Multi-Word Tokenizer that goes beyond word boundaries by representing frequent multi-word expressions as single tokens. MWTs produce a more compact and efficient tokenization that yields two benefits: (1) Increase in performance due to a greater coverage of input data given a fixed sequence length budget; (2) Faster and lighter inference due to the ability to reduce the sequence length with negligible drops in performance. Our results show that MWT is more robust across shorter sequence lengths, thus allowing for major speedups via early sequence truncation.

📄 PDF Abstract BibTeX arXiv:2402.09949

Code (1)

leonidasy/fast-vocabulary-transfer 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Broken Words, Broken Performance: Effect of Tokenization on Performance of LLMs

2025-12-26 · Sachin Pawar, Manoj Apte, Kshitij Jadhav, Girish Keshav Palshikar 외 arxiv

Tokenization is the first step in training any Large Language Model (LLM), where the text is split into a sequence of tokens as per the model's fixed vocabulary. This tokenization in LLMs is different from the traditiona…

MAGNET: Improving the Multilingual Fairness of Language Models with Adaptive Gradient-Based Tokenization

2024-07-11 · Orevaoghene Ahia, Sachin Kumar, Hila Gonen, Valentin Hofmann 외

In multilingual settings, non-Latin scripts and low-resource languages are usually disadvantaged in terms of language models' utility, efficiency, and cost. Specifically, previous studies have reported multiple modeling …

FairnessLanguage ModellingSegmentation

Analyzing Cognitive Plausibility of Subword Tokenization

2023-10-20 · Lisa Beinborn, Yuval Pinter

Subword tokenization has become the de-facto standard for tokenization, although comparative evaluations of subword vocabulary quality across languages are scarce. Existing evaluation studies focus on the effect of a tok…

AMBERT: A Pre-trained Language Model with Multi-Grained Tokenization

2020-08-27 · Findings (ACL) 2021 8 · Xinsong Zhang, Pengshuai Li, Hang Li

Pre-trained language models such as BERT have exhibited remarkable performances in many tasks in natural language understanding (NLU). The tokens in the models are usually fine-grained in the sense that for languages lik…

Language ModelingLanguage ModellingNatural Language Understanding

FLEXITOKENS: Flexible Tokenization for Evolving Language Models

2025-07-17 · Abraham Toluase Owodunni, Orevaoghene Ahia, Sachin Kumar

Language models (LMs) are challenging to adapt to new data distributions by simple finetuning. This is due to the rigidity of their subword tokenizers, which typically remain unchanged during adaptation. This inflexibili…