English
Related papers

Related papers: Separate Before You Compress: The WWHO Tokenizatio…

200 papers

While pixel-based language modeling aims to bypass the sub-word tokenization bottleneck by rendering text as images, recent multimodal variants such as DualGPT reintroduce text tokenizers to improve autoregressive performance. We…

Computation and Language · Computer Science 2026-02-10 Lucky Susanto , Musa Izzanardi Wijanarko , Khumaisa Nur'aini , Farid Adilazuarda , Alham Fikri Aji , Derry Tanti Wijaya

Tokenization is a critical preprocessing step in large language model pipelines, yet widely-used implementations remain CPU-bound and suboptimal for batch inference workflows on GPU. We present BlockBPE, a parallel GPU implementation of…

Computation and Language · Computer Science 2025-07-17 Amos You

Cutting-edge LLMs have emerged as powerful tools for multilingual communication and understanding. However, LLMs perform worse in Common Sense Reasoning (CSR) tasks when prompted in low-resource languages (LRLs) like Hindi or Swahili…

Computation and Language · Computer Science 2025-08-19 Tanay Nagar , Grigorii Khvatskii , Anna Sokol , Nitesh V. Chawla

Somali is a Cushitic language of the Horn of Africa with ~25 million speakers, yet no documented dedicated Somali pretraining corpus with a companion tokenizer and language-identification benchmark has been publicly released. Existing…

Computation and Language · Computer Science 2026-05-19 Khalid Yusuf Dahir

There are more than 7,000 languages around the world, and current Large Language Models (LLMs) only support hundreds of languages. Dictionary-based prompting methods can enhance translation on them, but most methods use all the available…

Computation and Language · Computer Science 2026-05-20 Hongyuan Lu , Zixuan Li , Zefan Zhang , Wai Lam

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

This paper evaluates the performance of several modern subword segmentation methods in a low-resource neural machine translation setting. We compare segmentations produced by applying BPE at the token or sentence level with…

Computation and Language · Computer Science 2024-05-17 Jonne Sälevä , Constantine Lignos

Token pruning has emerged as a mainstream approach for developing efficient Video Large Language Models (Video LLMs). This work revisits and advances the two predominant token-pruning paradigms: attention-based selection and…

Computer Vision and Pattern Recognition · Computer Science 2026-04-14 Shukang Yin , Sirui Zhao , Hanchao Wang , Baozhi Jia , Xianquan Wang , Chaoyou Fu , Enhong Chen

Quality Estimation (QE) aims to assess the quality of machine translation (MT) outputs without relying on reference translations, making it essential for real-world, large-scale MT evaluation. Large Language Models (LLMs) have shown…

Computation and Language · Computer Science 2026-02-10 Archchana Sindhujan , Girish A. Koushik , Shenbin Qian , Diptesh Kanojia , Constantin Orăsan

Text compression for large language model (LLM) systems is usually framed as token deletion, retrieval, summarization, or exact reconstruction. We study a more aggressive but explicitly lossy setting: compress text into compact codes that…

Machine Learning · Computer Science 2026-05-26 Natalia Trukhina , Vadim Vashkelis

End-to-end multilingual ASR has become more appealing because of several reasons such as simplifying the training and deployment process and positive performance transfer from high-resource to low-resource languages. However, scaling up the…

Computation and Language · Computer Science 2022-11-11 Andros Tjandra , Nayan Singhal , David Zhang , Ozlem Kalinli , Abdelrahman Mohamed , Duc Le , Michael L. Seltzer

Following the rationale of end-to-end modeling, CTC, RNN-T or encoder-decoder-attention models for automatic speech recognition (ASR) use graphemes or grapheme-based subword units based on e.g. byte-pair encoding (BPE). The mapping from…

Audio and Speech Processing · Electrical Eng. & Systems 2021-04-16 Mohammad Zeineldeen , Albert Zeyer , Wei Zhou , Thomas Ng , Ralf Schlüter , Hermann Ney

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

Large language models (LLMs) have transformed the software engineering landscape. Recently, numerous LLM-based agents have been developed to address real-world software issue fixing tasks. Despite their state-of-the-art performance, Despite…

Software Engineering · Computer Science 2026-03-10 Xin-Cheng Wen , Binbin Chen , Haoxuan Lan , Hang Yu , Peng Di , Cuiyun Gao

The paper presents an overview of the Spoken Language Translator (SLT) system's hybrid language-processing architecture, focussing on the way in which rule-based and statistical methods are combined to achieve robust and efficient…

cmp-lg · Computer Science 2008-02-03 Manny Rayner , David Carter

Large language models have drastically changed the prospects of AI by introducing technologies for more complex natural language processing. However, current methodologies to train such LLMs require extensive resources including but not…

Computation and Language · Computer Science 2026-04-27 Noel Elias , Homa Esfahanizadeh , Kaan Kale , Sriram Vishwanath , Muriel Medard

Recent advancements in technology have led to a boost in social media usage which has ultimately led to large amounts of user-generated data which also includes hateful and offensive speech. The language used in social media is often a…

Computation and Language · Computer Science 2025-05-27 Geet Shingi , Vedangi Wagh

Grapheme-to-phoneme (G2P) models are a key component in Automatic Speech Recognition (ASR) systems, such as the ASR system in Alexa, as they are used to generate pronunciations for out-of-vocabulary words that do not exist in the…

Computation and Language · Computer Science 2020-06-30 Alex Sokolov , Tracy Rohlin , Ariya Rastrow

The auto-regressive decoding of Large Language Models (LLMs) results in significant overheads in their hardware performance. While recent research has investigated various speculative decoding techniques for multi-token generation, these…

Machine Learning · Computer Science 2025-10-01 Hao Mark Chen , Wayne Luk , Ka Fai Cedric Yiu , Rui Li , Konstantin Mishchenko , Stylianos I. Venieris , Hongxiang Fan

Processing low-resource languages, such as Kiswahili, using machine learning is difficult due to lack of adequate training data. However, such low-resource languages are still important for human communication and are already in daily use…

Computation and Language · Computer Science 2025-01-17 Barack Wamkaya Wanjawa , Lawrence Muchemi , Evans Miriti