English
Related papers

Related papers: Compute Optimal Tokenization

200 papers

Large Audio Language Models (LALMs) demonstrate impressive performance across diverse tasks, ranging from speech recognition to general audio understanding. However, their scalability is limited by the quadratic complexity of attention and…

Audio and Speech Processing · Electrical Eng. & Systems 2025-11-27 Saurabhchand Bhati , Samuel Thomas , Hilde Kuehne , Rogerio Feris , James Glass

We introduce a simple modification to the embedding layer. The key change is to infuse token embeddings with information about their spelling. Models trained with these embeddings improve not only on spelling, but also across standard…

Machine Learning · Computer Science 2026-01-27 Markus N. Rabe , Judith Clymo , Zheren Dong

Because large language models are expensive to pretrain on different datasets, using smaller-scale experiments to decide on data is crucial for reducing costs. Which benchmarks and methods of making decisions from observed performance at…

Low-precision training is critical for optimizing the trade-off between model quality and training costs, necessitating the joint allocation of model size, dataset size, and numerical precision. While empirical scaling laws suggest that…

Machine Learning · Statistics 2026-02-27 Dechen Zhang , Xuan Tang , Yingyu Liang , Difan Zou

Tokenization is the act of breaking down text into smaller parts, or tokens, that are easier for machines to process. This is a key phase in machine translation (MT) models. Subword tokenization enhances this process by breaking down words…

Computation and Language · Computer Science 2025-05-23 Sudhansu Bala Das , Samujjal Choudhury , Tapas Kumar Mishra , Bidyut Kr. Patra

Tokenization is a fundamental component of large language models (LLMs), yet its influence on model scaling and performance is not fully explored. In this paper, we introduce Over-Tokenized Transformers, a novel framework that decouples…

Computation and Language · Computer Science 2025-05-26 Hongzhi Huang , Defa Zhu , Banggu Wu , Yutao Zeng , Ya Wang , Qiyang Min , Xun Zhou

Effective training of today's large language models (LLMs) depends on large batches and long sequences for throughput and accuracy. To handle variable-length sequences on hardware accelerators, it is common practice to introduce padding…

Computation and Language · Computer Science 2022-10-07 Mario Michael Krell , Matej Kosec , Sergio P. Perez , Andrew Fitzgibbon

The ColBERT model has recently been proposed as an effective BERT based ranker. By adopting a late interaction mechanism, a major advantage of ColBERT is that document representations can be precomputed in advance. However, the big downside…

Information Retrieval · Computer Science 2021-12-14 Carlos Lassance , Maroua Maachou , Joohee Park , Stéphane Clinchant

Large language model (LLM) scaling laws are empirical formulas that estimate changes in model quality as a result of increasing parameter count and training data. However, these formulas, including the popular Deepmind Chinchilla scaling…

Machine Learning · Computer Science 2025-04-15 Nikhil Sardana , Jacob Portes , Sasha Doubov , Jonathan Frankle

Transformer-based models have achieved dominant performance in numerous NLP tasks. Despite their remarkable successes, pre-trained transformers such as BERT suffer from a computationally expensive self-attention mechanism that interacts…

Computation and Language · Computer Science 2024-06-04 Jungmin Yun , Mihyeon Kim , Youngbin Kim

Tokenization-free hierarchical models are emerging as a promising alternative to traditional Large Language Models (LLMs), addressing inherent preprocessing issues such as vocabulary design complexity, out-of-vocabulary (OOV) errors, and…

Computation and Language · Computer Science 2026-05-29 Thang Dang , Akira Nakagawa , Kenichi Kobayashi , Koichi Shirahata

Unlike hybrid speech recognition systems where the use of tokens was restricted to phones, biphones or triphones the choice of tokens in the end-to-end ASR systems is derived from the text corpus of the training data. The use of…

Audio and Speech Processing · Electrical Eng. & Systems 2024-06-06 Sunil Kumar Kopparapu , Ashish Panda

Existing Machine Translation (MT) research often suggests a single, fixed set of hyperparameters for word segmentation models, symmetric Byte Pair Encoding (BPE), which applies the same number of merge operations (NMO) to train tokenizers…

Computation and Language · Computer Science 2026-02-16 Saumitra Yadav , Manish Shrivastava

Tokenizers provide the fundamental basis through which text is represented and processed by language models (LMs). Despite the importance of tokenization, its role in LM performance and behavior is poorly understood due to the challenge of…

Computation and Language · Computer Science 2025-12-25 Gül Sena Altıntaş , Malikeh Ehghaghi , Brian Lester , Fengyuan Liu , Wanru Zhao , Marco Ciccone , Colin Raffel

As we scale to more massive machine learning models, the frequent synchronization demands inherent in data-parallel approaches create significant slowdowns, posing a critical challenge to further scaling. Recent work develops an approach…

Machine Learning · Computer Science 2025-03-14 Zachary Charles , Gabriel Teston , Lucio Dery , Keith Rush , Nova Fallen , Zachary Garrett , Arthur Szlam , Arthur Douillard

Representations from transformer-based unidirectional language models are known to be effective at predicting brain responses to natural language. However, most studies comparing language models to brains have used GPT-2 or similarly sized…

Computation and Language · Computer Science 2024-01-31 Richard Antonello , Aditya Vaidya , Alexander G. Huth

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

ML-powered code generation aims to assist developers to write code in a more productive manner, by intelligently generating code blocks based on natural language prompts. Recently, large pretrained deep learning models have substantially…

Subword tokenization is a key part of many NLP pipelines. However, little is known about why some tokenizer and hyperparameter combinations lead to better downstream model performance than others. We propose that good tokenizers lead to…

Computation and Language · Computer Science 2023-06-30 Vilém Zouhar , Clara Meister , Juan Luis Gastaldi , Li Du , Mrinmaya Sachan , Ryan Cotterell

The laws of model size, data volume, computation and model performance have been extensively studied in the field of Natural Language Processing (NLP). However, the scaling laws in Optical Character Recognition (OCR) have not yet been…

Computer Vision and Pattern Recognition · Computer Science 2024-02-01 Miao Rang , Zhenni Bi , Chuanjian Liu , Yunhe Wang , Kai Han