English
Related papers

Related papers: Compute Optimal Tokenization

200 papers

Large language models (LLMs) have made remarkable advances in recent years, with scaling laws playing a critical role in this rapid progress. In this paper, we empirically investigate how a critical hyper-parameter, i.e., the global batch…

Computation and Language · Computer Science 2024-12-03 Xian Shuai , Yiding Wang , Yimeng Wu , Xin Jiang , Xiaozhe Ren

Traditional scaling laws in natural language processing suggest that increasing model size and training data enhances performance. However, recent studies reveal deviations, particularly in large language models, where performance…

Machine Learning · Computer Science 2025-07-16 Zhengyu Chen , Siqi Wang , Teng Xiao , Yudong Wang , Shiqi Chen , Xunliang Cai , Junxian He , Jingang Wang

Improvements in language model capabilities are often attributed to increasing model size or training data, but in some cases smaller models trained on curated data or with different architectural decisions can outperform larger ones…

Tokenization, a crucial initial step in natural language processing, is governed by several key parameters, such as the tokenization algorithm, vocabulary size, pre-tokenization strategy, inference strategy, and training data corpus. This…

Computation and Language · Computer Science 2025-06-17 Varshini Reddy , Craig W. Schmidt , Yuval Pinter , Chris Tanner

Empirically-determined scaling laws have been broadly successful in predicting the evolution of large machine learning models with training data and number of parameters. As a consequence, they have been useful for optimizing the allocation…

Computer Vision and Pattern Recognition · Computer Science 2024-07-26 Katherine L. Mentzer , Andrea Montanari

We explore optimally training protein language models, an area of significant interest in biological research where guidance on best practices is limited. Most models are trained with extensive compute resources until performance gains…

Machine Learning · Computer Science 2024-11-05 Xingyi Cheng , Bo Chen , Pan Li , Jing Gong , Jie Tang , Le Song

State-of-the-art language models are autoregressive and operate on subword units known as tokens. Specifically, one must encode the conditioning string into a list of tokens before passing to the language models for next-token prediction.…

Computation and Language · Computer Science 2024-07-09 Buu Phan , Marton Havasi , Matthew Muckley , Karen Ullrich

We investigate how different compression techniques -- such as weight and activation quantization, and weight sparsity -- affect the scaling behavior of large language models (LLMs) during pretraining. Building on previous work showing that…

Machine Learning · Computer Science 2025-02-27 Elias Frantar , Utku Evci , Wonpyo Park , Neil Houlsby , Dan Alistarh

We study the compute-optimal trade-off between model and training data set sizes for large neural networks. Our result suggests a linear relation similar to that supported by the empirical analysis of chinchilla. While that work studies…

Machine Learning · Computer Science 2023-10-20 Hong Jun Jeon , Benjamin Van Roy

Scaling laws for language models have often focused on finding the optimal model size and token count for training from scratch. However, achieving this optimal balance requires significant compute resources due to the extensive data…

Computation and Language · Computer Science 2025-04-01 Roseval Malaquias Junior , Ramon Pires , Thales Sales Almeida , Kenzo Sakiyama , Roseli A. F. Romero , Rodrigo Nogueira

The effectiveness of Neural Machine Translation (NMT) models largely depends on the vocabulary used at training; small vocabularies can lead to out-of-vocabulary problems -- large ones, to memory issues. Subword (SW) tokenization has been…

Computation and Language · Computer Science 2023-03-02 J. Pourmostafa Roshan Sharami , D. Shterionov , P. Spronck

Large language models are trained with tokenizers, and the resulting token distribution is highly imbalanced: a few words dominate the stream while most occur rarely. Recent practice favors ever-larger vocabularies, but it is unclear where…

Computation and Language · Computer Science 2025-12-01 Woojin Chung , Jeonghoon Kim

Recent work has shown that scaling large language models (LLMs) improves their alignment with human brain activity, yet it remains unclear what drives these gains and which representational properties are responsible. Although larger models…

Language models typically tokenize raw text into sequences of subword identifiers from a predefined vocabulary, a process inherently sensitive to typographical errors, length variations, and largely oblivious to the internal structure of…

Computation and Language · Computer Science 2024-10-07 Yekun Chai , Yewei Fang , Qiwei Peng , Xuhong Li

The success of Large Language Models (LLMs) has established that scaling compute, through joint increases in model capacity and dataset size, is the primary driver of performance in modern machine learning. While machine learning has long…

High Energy Physics - Experiment · Physics 2026-02-18 Matthias Vigl , Nicole Hartman , Michael Kagan , Lukas Heinrich

Building effective tokenizers for multilingual Large Language Models (LLMs) requires careful control over language-specific data mixtures. While a tokenizer's compression performance critically affects the efficiency of LLM training and…

Computation and Language · Computer Science 2026-01-21 Inho Won , Hangyeol Yoo , Minkyung Cho , Jungyeul Park , Hoyun Song , KyungTae Lim

The scaling law is becoming a fundamental law in many machine learning areas. That is, test error falls off with the power law when increasing training data, model size, and computing resource. However, whether this law is suitable for the…

Software Engineering · Computer Science 2024-02-21 Jiayi Lin , Hande Dong , Yutao Xie , Lei Zhang

Large language model pre-training has become increasingly expensive, with most practitioners relying on scaling laws to allocate compute budgets for model size and training tokens, commonly referred to as Compute-Optimal or Chinchilla…

Machine Learning · Computer Science 2024-05-03 Zhen Guo

Deep neural networks generally involve some layers with mil- lions of parameters, making them difficult to be deployed and updated on devices with limited resources such as mobile phones and other smart embedded systems. In this paper, we…

Computer Vision and Pattern Recognition · Computer Science 2016-08-29 Xing Wang , Jie Liang

The prevalent use of Byte Pair Encoding (BPE) in Large Language Models (LLMs) facilitates robust handling of subword units and avoids issues of out-of-vocabulary words. Despite its success, a critical challenge persists: long tokens, rich…

Computation and Language · Computer Science 2024-11-11 Haoran Lian , Yizhe Xiong , Zijia Lin , Jianwei Niu , Shasha Mo , Hui Chen , Peng Liu , Guiguang Ding