English
Related papers

Related papers: Significance-Gain Pair Encoding for LLMs: A Statis…

200 papers

Phones and their context-dependent variants have been the standard modeling units for conventional speech recognition systems, while characters and subwords have demonstrated their effectiveness for end-to-end recognition systems. We…

Audio and Speech Processing · Electrical Eng. & Systems 2021-06-23 Weiran Wang , Guangsen Wang , Aadyot Bhatnagar , Yingbo Zhou , Caiming Xiong , Richard Socher

Subword tokenization is an essential part of modern large language models (LLMs), yet its specific contributions to training efficiency and model performance remain poorly understood. In this work, we decouple the effects of subword…

Computation and Language · Computer Science 2026-05-15 Théo Gigant , Bowen Peng , Jeffrey Quesnelle

Large language model (LLM) tokenizers act as structured compressors: by mapping text to discrete token sequences, they determine token count (and thus compute and context usage) and the statistical structure seen by downstream models.…

Information Theory · Computer Science 2026-01-15 Mete Erdogan , Abhiram Gorle , Shubham Chandak , Mert Pilanci , Tsachy Weissman

Recent dynamic tokenisation methods operate directly on bytes and pool their latent representations into patches. This bears similarities to computational models of word segmentation that determine lexical boundaries using spikes in an…

Computation and Language · Computer Science 2025-06-24 Zébulon Goriely , Suchir Salhan , Pietro Lesci , Julius Cheng , Paula Buttery

Multilingual modelling can improve machine translation for low-resource languages, partly through shared subword representations. This paper studies the role of subword segmentation in cross-lingual transfer. We systematically compare the…

Computation and Language · Computer Science 2024-04-01 Francois Meyer , Jan Buys

Mixture-of-Experts large language models (MoE-LLMs) marks a significant step forward of language models, however, they encounter two critical challenges in practice: 1) expert parameters lead to considerable memory consumption and loading…

Machine Learning · Computer Science 2025-02-25 Wei Huang , Yue Liao , Jianhui Liu , Ruifei He , Haoru Tan , Shiming Zhang , Hongsheng Li , Si Liu , Xiaojuan Qi

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

Neural sentence embedding models for dense retrieval typically rely on binary relevance labels, treating query-document pairs as either relevant or irrelevant. However, real-world relevance often exists on a continuum, and recent advances…

Information Retrieval · Computer Science 2025-08-12 Christos Tsirigotis , Vaibhav Adlakha , Joao Monteiro , Aaron Courville , Perouz Taslakian

Scaling laws enable the optimal selection of data amount and language model size, yet the impact of the data unit, the token, on this relationship remains underexplored. In this work, we systematically investigate how the information…

Computation and Language · Computer Science 2026-05-27 Tomasz Limisiewicz , Artidoro Pagnoni , Srini Iyer , Mike Lewis , Sachin Mehta , Alisa Liu , Margaret Li , Gargi Ghosh , Luke Zettlemoyer

Subword tokenization has become the prevailing standard in the field of natural language processing (NLP) over recent years, primarily due to the widespread utilization of pre-trained language models. This shift began with Byte-Pair…

Computation and Language · Computer Science 2024-06-11 Yanis Labrak , Adrien Bazoge , Beatrice Daille , Mickael Rouvier , Richard Dufour

Large language models (LLMs) face significant token efficiency bottlenecks in code generation and logical reasoning tasks, a challenge that directly impacts inference cost and model interpretability. This paper proposes a formal framework…

Artificial Intelligence · Computer Science 2025-02-03 Lumen AI , Tengzhou No. 1 Middle School , Shihao Ji , Zihui Song , Fucheng Zhong , Jisen Jia , Zhaobo Wu , Zheyi Cao , Tianhao Xu

Subword tokenization schemes are the dominant technique used in current NLP models. However, such schemes can be rigid and tokenizers built on one corpus do not adapt well to other parallel corpora. It has also been observed that in…

Computation and Language · Computer Science 2023-06-29 Makesh Narsimhan Sreedhar , Xiangpeng Wan , Yu Cheng , Junjie Hu

An important assumption that comes with using LLMs on psycholinguistic data has gone unverified. LLM-based predictions are based on subword tokenization, not decomposition of words into morphemes. Does that matter? We carefully test this by…

Computation and Language · Computer Science 2023-10-30 Sathvik Nair , Philip Resnik

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

As the ever-increasing token limits of large language models (LLMs) have enabled long context as input, prompting with single data samples might no longer an efficient way. A straightforward strategy improving efficiency is to batch data…

Computation and Language · Computer Science 2024-07-16 Jianzhe Lin , Maurice Diesendruck , Liang Du , Robin Abraham

In this paper, we formalize practical byte pair encoding tokenization as it is used in large language models and other NLP systems, in particular we formally define and investigate the semantics of the SentencePiece and HuggingFace…

Formal Languages and Automata Theory · Computer Science 2023-09-19 Martin Berglund , Brink van der Merwe

We present a large-scale comparative study of 242 Latin and Cyrillic-script languages using subword-based methodologies. By constructing 'glottosets' from Wikipedia lexicons, we introduce a framework for simultaneous cross-linguistic…

Computation and Language · Computer Science 2026-01-27 Iaroslav Chelombitko , Mika Hämäläinen , Aleksey Komissarov

In recent years, language models have become increasingly larger and more complex. However, the input representations for these models continue to rely on simple and greedy subword tokenization methods. In this paper, we propose a novel…

Computation and Language · Computer Science 2023-06-14 David Samuel , Lilja Øvrelid

Discretizing speech into tokens and generating them by a decoder-only model have been a promising direction for text-to-speech (TTS) and spoken language modeling (SLM). To shorten the sequence length of speech tokens, acoustic byte-pair…

Sound · Computer Science 2024-10-30 Bohan Li , Feiyu Shen , Yiwei Guo , Shuai Wang , Xie Chen , Kai Yu

State-of-the-art natural language generation evaluation metrics are based on black-box language models. Hence, recent works consider their explainability with the goals of better understandability for humans and better metric analysis,…

Computation and Language · Computer Science 2024-02-20 Christoph Leiter , Hoa Nguyen , Steffen Eger