English
Related papers

Related papers: Language Models over Canonical Byte-Pair Encodings

200 papers

Modern tokenizers employ deterministic algorithms to map text into a single "canonical" token sequence, yet the same string can be encoded as many non-canonical tokenizations using the tokenizer vocabulary. In this work, we investigate the…

Computation and Language · Computer Science 2026-02-04 Brian Siyuan Zheng , Alisa Liu , Orevaoghene Ahia , Jonathan Hayase , Yejin Choi , Noah A. Smith

State-of-the-art language models are autoregressive and operate on subword units known as tokens. Specifically, one must encode the conditioning string into a list of tokens before passing to the language models for next-token prediction.…

Computation and Language · Computer Science 2024-07-09 Buu Phan , Marton Havasi , Matthew Muckley , Karen Ullrich

Large Language Models (LLMs) are typically shipped with tokenizers that deterministically encode text into so-called canonical token sequences, to which the LLMs assign probability values. One common assumption is that the probability of a…

Computation and Language · Computer Science 2025-06-09 Renato Lui Geh , Honghua Zhang , Kareem Ahmed , Benjie Wang , Guy Van den Broeck

Language models typically tokenize text into subwords, using a deterministic, hand-engineered heuristic of combining characters into longer surface-level strings such as 'ing' or whole words. Recent literature has repeatedly shown the…

Computation and Language · Computer Science 2023-10-19 Avijit Thawani , Saurabh Ghanekar , Xiaoyuan Zhu , Jay Pujara

Modern language models are internally -- and mathematically -- distributions over $\it{token}$ strings rather than $\it{character}$ strings, posing numerous challenges for programmers building user applications on top of them. For example,…

Computation and Language · Computer Science 2025-06-11 Tim Vieira , Ben LeBrun , Mario Giulianelli , Juan Luis Gastaldi , Brian DuSell , John Terilla , Timothy J. O'Donnell , Ryan Cotterell

Modern language models define distributions over strings, but downstream tasks often require different output formats. For instance, a model that generates byte-pair strings does not directly produce word-level predictions, and a DNA model…

Computation and Language · Computer Science 2026-03-09 Vésteinn Snæbjarnarson , Samuel Kiegeland , Tianyu Liu , Reda Boumasmoud , Ryan Cotterell , Tim Vieira

Modern language models are typically trained over subword sequences, but ultimately define probabilities over character-strings. Ideally, the choice of the tokeniser -- which maps character-strings to subwords -- should not affect the…

Computation and Language · Computer Science 2025-06-04 Pietro Lesci , Clara Meister , Thomas Hofmann , Andreas Vlachos , Tiago Pimentel

Tokenization - the practice of converting strings of characters from an alphabet into sequences of tokens over a vocabulary - is a critical step in the NLP pipeline. The use of token representations is widely credited with increased model…

Computation and Language · Computer Science 2025-04-04 Juan Luis Gastaldi , John Terilla , Luca Malagutti , Brian DuSell , Tim Vieira , Ryan Cotterell

Tokenization is the first step in modern neural language model pipelines where an input text is converted to a sequence of subword tokens. We introduce from first principles a finite-state transduction framework which can efficiently encode…

Computation and Language · Computer Science 2024-10-22 Marco Cognetta , Naoaki Okazaki

Tokenization underlies every large language model, yet it remains an under-theorized and inconsistently designed component. Common subword approaches such as Byte Pair Encoding (BPE) offer scalability but often misalign with linguistic…

Computation and Language · Computer Science 2026-01-27 Sawsan Alqahtani , Mir Tafseer Nayeem , Md Tahmid Rahman Laskar , Tasnim Mohiuddin , M Saiful Bari

Neural language models typically tokenise input text into sub-word units to achieve an open vocabulary. The standard approach is to use a single canonical tokenisation at both train and test time. We suggest that this approach is…

Computation and Language · Computer Science 2021-09-22 Kris Cao , Laura Rimell

In this paper, we investigate the output token probability information in the output embedding of language models. We find an approximate common log-linear encoding of output token probabilities within the output embedding vectors and…

Computation and Language · Computer Science 2024-12-12 Hakaze Cho , Yoshihiro Sakai , Kenshiro Tanaka , Mariko Kato , Naoya Inoue

Tokenization imposes a fixed granularity on the input text, freezing how a language model operates on data and how far in the future it predicts. Byte Pair Encoding (BPE) and similar schemes split text once, build a static vocabulary, and…

Computation and Language · Computer Science 2025-06-18 Mathurin Videau , Badr Youbi Idrissi , Alessandro Leite , Marc Schoenauer , Olivier Teytaud , David Lopez-Paz

Recent language models have shown impressive multilingual performance, even when not explicitly trained for it. Despite this, there are concerns about the quality of their outputs across different languages. In this paper, we show how…

Computation and Language · Computer Science 2023-10-23 Aleksandar Petrov , Emanuele La Malfa , Philip H. S. Torr , Adel Bibi

The choice of tokenizer can profoundly impact language model performance, yet accessible and reliable evaluations of tokenizer quality remain an open challenge. Inspired by scaling consistency, we show that smaller models can accurately…

Computation and Language · Computer Science 2025-06-04 Jonas F. Lotz , António V. Lopes , Stephan Peitz , Hendra Setiawan , Leonardo Emili

Language models typically tokenize raw text into sequences of subword identifiers from a predefined vocabulary, a process inherently sensitive to typographical errors, length variations, and largely oblivious to the internal structure of…

Computation and Language · Computer Science 2024-10-07 Yekun Chai , Yewei Fang , Qiwei Peng , Xuhong Li

Tokenization is associated with many poorly understood shortcomings in language models (LMs), yet remains an important component for long sequence scaling purposes. This work studies how tokenization impacts model performance by analyzing…

Computation and Language · Computer Science 2025-04-15 Buu Phan , Brandon Amos , Itai Gat , Marton Havasi , Matthew Muckley , Karen Ullrich

Subword tokenization methods, such as Byte-Pair Encoding (BPE), significantly impact the performance and efficiency of large language models (LLMs). The standard approach involves training a general-purpose tokenizer that uniformly…

Computation and Language · Computer Science 2026-01-30 Vijini Liyanage , François Yvon

Language models are widely used in computational psycholinguistics to test theories that relate the negative log probability (the surprisal) of a region of interest (a substring of characters) under a language model to its cognitive cost…

Computation and Language · Computer Science 2024-12-09 Mario Giulianelli , Luca Malagutti , Juan Luis Gastaldi , Brian DuSell , Tim Vieira , Ryan Cotterell

Modern language models mostly take sub-words as input, a design that balances the trade-off between vocabulary size, number of parameters, and performance. However, sub-word tokenization still has disadvantages like not being robust to…

Computation and Language · Computer Science 2022-11-24 Chu-Tak Lee , Qipeng Guo , Xipeng Qiu
‹ Prev 1 2 3 10 Next ›