English
Related papers

Related papers: Neural Machine Translation with Byte-Level Subword…

200 papers

Subword regularization methods such as BPE dropout are typically applied only during fine-tuning, while pretraining is usually done with deterministic tokenization. This creates a potential segmentation mismatch between pretraining and…

Computation and Language · Computer Science 2026-05-14 Ruan Visser , Trienko Grobler , Marcel Dunaiski

Most widely-used pre-trained language models operate on sequences of tokens corresponding to word or subword units. By comparison, token-free models that operate directly on raw text (bytes or characters) have many benefits: they can…

Computation and Language · Computer Science 2022-03-09 Linting Xue , Aditya Barua , Noah Constant , Rami Al-Rfou , Sharan Narang , Mihir Kale , Adam Roberts , Colin Raffel

In this study, we propose a simple and effective preprocessing method for subword segmentation based on a data compression algorithm. Compression-based subword segmentation has recently attracted significant attention as a preprocessing…

Computation and Language · Computer Science 2023-03-02 Keita Nonaka , Kazutaka Yamanouchi , Tomohiro I , Tsuyoshi Okita , Kazutaka Shimada , Hiroshi Sakamoto

Most studies on word-level Quality Estimation (QE) of machine translation focus on language-specific models. The obvious disadvantages of these approaches are the need for labelled data for each language pair and the high cost required to…

Computation and Language · Computer Science 2021-06-02 Tharindu Ranasinghe , Constantin Orasan , Ruslan Mitkov

Morphologically-rich polysynthetic languages present a challenge for NLP systems due to data sparsity, and a common strategy to handle this issue is to apply subword segmentation. We investigate a wide variety of supervised and unsupervised…

Computation and Language · Computer Science 2022-03-18 Manuel Mager , Arturo Oncevay , Elisabeth Mager , Katharina Kann , Ngoc Thang Vu

Binary Neural Network (BNN) represents convolution weights with 1-bit values, which enhances the efficiency of storage and computation. This paper is motivated by a previously revealed phenomenon that the binary kernels in successful BNNs…

Computer Vision and Pattern Recognition · Computer Science 2023-03-28 Yikai Wang , Wenbing Huang , Yinpeng Dong , Fuchun Sun , Anbang Yao

The best performing transformer-based language models use subword tokenization techniques, such as Byte-Pair-Encoding (BPE). However, these approaches often overlook linguistic principles, such as morphological segmentation, which we…

Computation and Language · Computer Science 2025-04-03 Mikkel Wildner Kildeberg , Emil Allerslev Schledermann , Nicolaj Larsen , Rob van der Goot

Multilingual training of neural machine translation (NMT) systems has led to impressive accuracy improvements on low-resource languages. However, there are still significant challenges in efficiently learning word representations in the…

Computation and Language · Computer Science 2019-02-12 Xinyi Wang , Hieu Pham , Philip Arthur , Graham Neubig

Translating characters instead of words or word-fragments has the potential to simplify the processing pipeline for neural machine translation (NMT), and improve results by eliminating hyper-parameters and manual feature engineering.…

Computation and Language · Computer Science 2018-08-30 Colin Cherry , George Foster , Ankur Bapna , Orhan Firat , Wolfgang Macherey

Word embedding is central to neural machine translation (NMT), which has attracted intensive research interest in recent years. In NMT, the source embedding plays the role of the entrance while the target embedding acts as the terminal.…

Computation and Language · Computer Science 2019-06-10 Xuebo Liu , Derek F. Wong , Yang Liu , Lidia S. Chao , Tong Xiao , Jingbo Zhu

Byte-Pair Encoding (BPE) is a popular algorithm used for tokenizing data in NLP, despite being devised initially as a compression method. BPE appears to be a greedy algorithm at face value, but the underlying optimization problem that BPE…

Computation and Language · Computer Science 2024-09-04 Vilém Zouhar , Clara Meister , Juan Luis Gastaldi , Li Du , Tim Vieira , Mrinmaya Sachan , Ryan Cotterell

Large language models (LLMs) have achieved remarkable success across various natural language processing tasks. However, most LLM models use traditional tokenizers like BPE and SentencePiece, which fail to capture the finer nuances of a…

Computation and Language · Computer Science 2025-05-26 Pramit Bhattacharyya , Arnab Bhattacharya

Language modeling is a fundamental task in natural language processing, which has been thoroughly explored with various architectures and hyperparameters. However, few studies focus on the effect of sub-word segmentation on the performance…

Computation and Language · Computer Science 2023-10-30 Jue Hou , Anisia Katinskaia , Anh-Duc Vu , Roman Yangarber

Neural machine translation (NMT) has a drawback in that can generate only high-frequency words owing to the computational costs of the softmax function in the output layer. In Japanese-English NMT, Japanese predicate conjugation causes an…

Computation and Language · Computer Science 2018-05-28 Michiki Kurosawa , Yukio Matsumura , Hayahide Yamagishi , Mamoru Komachi

We present a method to compress the final linear layer of language models, reducing memory usage by up to 3.4x without significant performance loss. By grouping tokens based on Byte Pair Encoding (BPE) merges, we prevent materialization of…

Computation and Language · Computer Science 2024-11-12 Sreeram Vennam , Anish Joishy , Ponnurangam Kumaraguru

Subword-level models have been the dominant paradigm in NLP. However, character-level models have the benefit of seeing each character individually, providing the model with more detailed information that ultimately could lead to better…

Computation and Language · Computer Science 2022-12-05 Lukas Edman , Antonio Toral , Gertjan van Noord

In this paper, we present an end-to-end automatic speech recognition system, which successfully employs subword units in a hybrid CTC-Attention based system. The subword units are obtained by the byte-pair encoding (BPE) compression…

Audio and Speech Processing · Electrical Eng. & Systems 2018-09-07 Zhangyu Xiao , Zhijian Ou , Wei Chu , Hui Lin

Automated malware analysis increasingly relies on machine learning, yet most existing methods remain task-specific and depend on handcrafted features or narrowly scoped models. Recent developments in binary-level foundation models suggest a…

Cryptography and Security · Computer Science 2026-05-19 Saastha Vasan , Yuzhou Nie , Kaie Chen , Yigitcan Kaya , Hojjat Aghakhani , Roman Vasilenko , Wenbo Guo , Christopher Kruegel , Giovanni Vigna

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

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