English
Related papers

Related papers: Rethinking Tokenization: Crafting Better Tokenizer…

200 papers

A better understanding of the emergent computation and problem-solving capabilities of recent large language models is of paramount importance to further improve them and broaden their applicability. This work investigates how a language…

Artificial Intelligence · Computer Science 2024-08-05 Davide Maltoni , Matteo Ferrara

Choosing an appropriate tokenization scheme is often a bottleneck in low-resource cross-lingual transfer. To understand the downstream implications of text representation choices, we perform a comparative analysis on language models having…

Computation and Language · Computer Science 2023-10-13 Md Mushfiqur Rahman , Fardin Ahsan Sakib , Fahim Faisal , Antonios Anastasopoulos

Almost all existing machine translation models are built on top of character-based vocabularies: characters, subwords or words. Rare characters from noisy text or character-rich languages such as Japanese and Chinese however can…

Computation and Language · Computer Science 2019-12-09 Changhan Wang , Kyunghyun Cho , Jiatao Gu

This paper investigates the effect of tokenizers on the downstream performance of pretrained language models (PLMs) in scriptio continua languages where no explicit spaces exist between words, using Japanese as a case study. The tokenizer…

Computation and Language · Computer Science 2023-06-19 Takuro Fujii , Koki Shibata , Atsuki Yamaguchi , Terufumi Morishita , Yasuhiro Sogawa

Tokenization is a pivotal design choice for neural language modeling in morphologically rich languages (MRLs) such as Turkish, where productive agglutination challenges both vocabulary efficiency and morphological fidelity. Prior studies…

Computation and Language · Computer Science 2026-02-09 Duygu Altinok

The popular subword tokenizers of current language models, such as Byte-Pair Encoding (BPE), are known not to respect morpheme boundaries, which affects the downstream performance of the models. While many improved tokenization algorithms…

Computation and Language · Computer Science 2024-04-23 Khuyagbaatar Batsuren , Ekaterina Vylomova , Verna Dankers , Tsetsuukhei Delgerbaatar , Omri Uzan , Yuval Pinter , Gábor Bella

Standard Byte-Pair Encoding (BPE) tokenization compresses text by pairing a learned token vocabulary with a detailed merge list. Recent work has shown that this merge list exposes a potential attack surface for extracting information about…

Computation and Language · Computer Science 2025-08-12 Tomohiro Sawada , Kartik Goyal

Subword tokenization introduces a computational layer in language models where many distinct token sequences decode to the same surface form and preserve meaning, yet induce different internal computations. Despite this non-uniqueness,…

Computation and Language · Computer Science 2026-01-14 Adrian Cosma , Stefan Ruseti , Emilian Radoi , Mihai Dascalu

Subword tokenization is a key design choice for modern language models, including large language models (LLMs), with byte- and character-level BPE serving as a widely used baseline. Standard BPE selects merges by raw pair frequency, which…

Computation and Language · Computer Science 2026-03-23 Azam Nouri

Model ensembling is a technique to combine the predicted distributions of two or more models, often leading to improved robustness and performance. For ensembling in text generation, the next token's probability distribution is derived from…

Computation and Language · Computer Science 2025-03-03 Rachel Wicks , Kartik Ravisankar , Xinchen Yang , Philipp Koehn , Matt Post

Tokenization is fundamental to pretrained language models (PLMs). Existing tokenization methods for Chinese PLMs typically treat each character as an indivisible token. However, they ignore the unique feature of the Chinese writing system…

Computation and Language · Computer Science 2023-02-16 Chenglei Si , Zhengyan Zhang , Yingfa Chen , Fanchao Qi , Xiaozhi Wang , Zhiyuan Liu , Yasheng Wang , Qun Liu , Maosong Sun

The impact of subword tokenization on language model performance is well-documented for perplexity, with finer granularity consistently reducing this intrinsic metric. However, research on how different tokenization schemes affect a model's…

Computation and Language · Computer Science 2025-08-12 Nishant Luitel , Nirajan Bekoju , Anand Kumar Sah , Subarna Shakya

Although recent advances in scaling large language models (LLMs) have resulted in improvements on many NLP tasks, it remains unclear whether these models trained primarily with general web text are the right tool in highly specialized,…

Recent large language models (LLM) exhibit sub-optimal performance on low-resource languages, as the training data of these models is usually dominated by English and other high-resource languages. Furthermore, it is challenging to train…

Computation and Language · Computer Science 2023-12-18 Zoltan Csaki , Pian Pawakapan , Urmish Thakker , Qiantong Xu

Recent advancements in large language models(LLMs), such as GPT-4 and GPT-4o, have shown exceptional performance, especially in languages with abundant resources like English, thanks to extensive datasets that ensure robust training.…

Computation and Language · Computer Science 2024-11-15 Jin Yang , Zhiqiang Wang , Yanbin Lin , Zunduo Zhao

Generative language models are usually pretrained on large text corpus via predicting the next token (i.e., sub-word/word/phrase) given the previous ones. Recent works have demonstrated the impressive performance of large generative…

Computation and Language · Computer Science 2026-01-16 Zhenpeng Su , Xing Wu , Xue Bai , Zijia Lin , Hui Chen , Guiguang Ding , Wei Zhou , Songlin Hu

Tokenization is an important first step in Natural Language Processing (NLP) pipelines because it decides how models learn and represent linguistic information. However, current subword tokenizers like SentencePiece or HuggingFace BPE are…

Computation and Language · Computer Science 2025-11-10 Firoj Ahmmed Patwary , Abdullah Al Noman

A distinction is often drawn between a model's ability to predict a label for an evaluation sample that is directly memorised from highly similar training samples versus an ability to predict the label via some method of generalisation. In…

Computation and Language · Computer Science 2023-11-22 Tim Hartill , Joshua Bensemann , Michael Witbrock , Patricia J. Riddle

Subword tokenization methods like Byte Pair Encoding (BPE) are widely used in large language models due to their balance of vocabulary compactness and representational power. However, they suffer from inefficiencies in representing rare…

Computation and Language · Computer Science 2025-10-20 Rares Dolga , Lucas Maystre , Tudor Berariu , David Barber

All languages are equal; when it comes to tokenization, some are more equal than others. Tokens are the hidden currency that dictate the cost and latency of access to contemporary LLMs. However, many languages written in non-Latin scripts…

Computation and Language · Computer Science 2026-04-21 Maitrey Mehta , Nishant Subramani , Zhichao Xu , Ashim Gupta , Vivek Srikumar