English
Related papers

Related papers: SuperBPE: Space Travel for Language Models

200 papers

Most neural machine translation systems are built upon subword units extracted by methods such as Byte-Pair Encoding (BPE) or wordpiece. However, the choice of number of merge operations is generally made by following existing recipes. In…

Computation and Language · Computer Science 2019-06-26 Shuoyang Ding , Adithya Renduchintala , Kevin Duh

Subword tokenization critically affects Natural Language Processing (NLP) performance, yet its behavior in morphologically rich and low-resource language families remains under-explored. This study systematically compares three subword…

Computation and Language · Computer Science 2026-03-31 Nuo Xu , Ahrii Kim

Large pretrained language models (LMs) have become the central building block of many NLP applications. Training these models requires ever more computational resources and most of the existing models are trained on English text only. It is…

Computation and Language · Computer Science 2022-09-13 Benjamin Minixhofer , Fabian Paischer , Navid Rekabsaz

As opposed to general English, many concepts in biomedical terminology have been designed in recent history by biomedical professionals with the goal of being precise and concise. This is often achieved by concatenating meaningful…

Computation and Language · Computer Science 2023-07-11 Bernal Jiménez Gutiérrez , Huan Sun , Yu Su

Tokenization methods like Byte-Pair Encoding (BPE) enhance computational efficiency in large language models (LLMs) but often obscure internal character structures within tokens. This limitation hinders LLMs' ability to predict precise…

Computation and Language · Computer Science 2025-06-10 Zhu Xu , Zhiqiang Zhao , Zihan Zhang , Yuchi Liu , Quanwei Shen , Fei Liu , Yu Kuang , Jian He , Conglin Liu

Subword tokenization requires balancing computational efficiency and vocabulary coverage, which often leads to suboptimal performance on languages and scripts not prioritized during training. We propose to augment pretrained language models…

Computation and Language · Computer Science 2025-08-12 Jonas F. Lotz , Hendra Setiawan , Stephan Peitz , Yova Kementchedjhieva

As a cornerstone in language modeling, tokenization involves segmenting text inputs into pre-defined atomic units. Conventional statistical tokenizers often disrupt constituent boundaries within words, thereby corrupting semantic…

Computation and Language · Computer Science 2025-07-11 Qingyang Zhu , Xiang Hu , Pengyu Ji , Wei Wu , Kewei Tu

The Backpack is a Transformer alternative shown to improve interpretability in English language modeling by decomposing predictions into a weighted sum of token sense components. However, Backpacks' reliance on token-defined meaning raises…

Computation and Language · Computer Science 2023-10-20 Hao Sun , John Hewitt

Computing next-token likelihood ratios between two language models (LMs) is a standard task in training paradigms such as knowledge distillation. Since this requires both models to share the same probability space, it becomes challenging…

Computation and Language · Computer Science 2026-05-07 Buu Phan , Ashish Khisti , Karen Ullrich

Byte-Pair Encoding (BPE) is a widely used method for subword tokenization, with origins in grammar-based text compression. It is employed in a variety of language processing tasks such as machine translation or large language model (LLM)…

Data Structures and Algorithms · Computer Science 2024-11-14 László Kozma , Johannes Voderholzer

Subword segmentation is widely used to address the open vocabulary problem in machine translation. The dominant approach to subword segmentation is Byte Pair Encoding (BPE), which keeps the most frequent words intact while splitting the…

Computation and Language · Computer Science 2020-05-05 Ivan Provilkov , Dmitrii Emelianenko , Elena Voita

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 tokenization algorithm…

Computation and Language · Computer Science 2023-10-23 Lisa Beinborn , Yuval Pinter

The number of tokens it takes to encode parallel text in different languages is known to vary. These disparities are called token premiums. Having high token premiums leads to less throughput during training and increases costs at…

Computation and Language · Computer Science 2025-10-28 Catherine Arnett , Tyler A. Chang , Stella Biderman , Benjamin K. Bergen

Recent work on tokenizer-free multilingual pretrained models show promising results in improving cross-lingual transfer and reducing engineering overhead (Clark et al., 2022; Xue et al., 2022). However, these works mainly focus on reporting…

Computation and Language · Computer Science 2022-10-14 Jimin Sun , Patrick Fernandes , Xinyi Wang , Graham Neubig

While model architecture and training objectives are well-studied, tokenization, particularly in multilingual contexts, remains a relatively neglected aspect of Large Language Model (LLM) development. Existing tokenizers often exhibit high…

The Byte Pair Encoding algorithm can be safely batched to merge hundreds of pairs of tokens at a time when building up a tokenizer's vocabulary. This technique combined with reducing the memory footprint of text used in vocabulary training…

Computation and Language · Computer Science 2024-08-12 Alexander P. Morgan

Multimodal Large Language Models have made significant strides in integrating visual and textual information, yet they often struggle with effectively aligning these modalities. We introduce a novel image tokenizer that bridges this gap by…

Artificial Intelligence · Computer Science 2025-03-11 Wanpeng Zhang , Zilong Xie , Yicheng Feng , Yijiang Li , Xingrun Xing , Sipeng Zheng , Zongqing Lu

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

Sub-word segmentation is an essential pre-processing step for Neural Machine Translation (NMT). Existing work has shown that neural sub-word segmenters are better than Byte-Pair Encoding (BPE), however, they are inefficient as they require…

Computation and Language · Computer Science 2023-08-01 Haiyue Song , Raj Dabre , Chenhui Chu , Sadao Kurohashi , Eiichiro Sumita

Training data memorization in NLP can both be beneficial (e.g., closed-book QA) and undesirable (personal data extraction). In any case, successful model training requires a non-trivial amount of memorization to store word spellings,…

Computation and Language · Computer Science 2021-12-03 Eugene Kharitonov , Marco Baroni , Dieuwke Hupkes