English
Related papers

Related papers: Pre-trained Models Perform the Best When Token Dis…

200 papers

The word embedding space in neural models is skewed, and correcting this can improve task performance. We point out that most approaches for modeling, correcting, and measuring the symmetry of an embedding space implicitly assume that the…

Computation and Language · Computer Science 2024-11-04 Sho Yokoi , Han Bao , Hiroto Kurita , Hidetoshi Shimodaira

The choice of tokenizer can profoundly impact language model performance, yet accessible and reliable evaluations of tokenizer quality remain an open challenge. Inspired by scaling consistency, we show that smaller models can accurately…

Computation and Language · Computer Science 2025-06-04 Jonas F. Lotz , António V. Lopes , Stephan Peitz , Hendra Setiawan , Leonardo Emili

We inspect the deductive connection between the neural scaling law and Zipf's law -- two statements discussed in machine learning and quantitative linguistics. The neural scaling law describes how the cross entropy rate of a foundation…

Information Theory · Computer Science 2025-12-23 Łukasz Dębowski

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

This paper studies the limits of language models' statistical learning in the context of Zipf's law. First, we demonstrate that Zipf-law token distribution emerges irrespective of the chosen tokenization. Second, we show that Zipf…

Computation and Language · Computer Science 2022-11-22 Elizaveta Zhemchuzhina , Nikolai Filippov , Ivan P. Yamshchikov

In this study, we investigate whether speech symbols, learned through deep learning, follow Zipf's law, akin to natural language symbols. Zipf's law is an empirical law that delineates the frequency distribution of words, forming…

Computation and Language · Computer Science 2023-09-19 Shinnosuke Takamichi , Hiroki Maeda , Joonyong Park , Daisuke Saito , Hiroshi Saruwatari

Recent works have highlighted optimization difficulties faced by gradient descent in training the first and last layers of transformer-based language models, which are overcome by optimizers such as Adam. These works suggest that the…

Machine Learning · Computer Science 2025-05-27 Frederik Kunstner , Francis Bach

The formation of sentences is a highly structured and history-dependent process. The probability of using a specific word in a sentence strongly depends on the 'history' of word-usage earlier in that sentence. We study a simple…

Physics and Society · Physics 2015-05-28 Stefan Thurner , Rudolf Hanel , Bo Liu , Bernat Corominas-Murtra

Tokenization is a crucial step in processing protein sequences for machine learning models, as proteins are complex sequences of amino acids that require meaningful segmentation to capture their functional and structural properties.…

Computation and Language · Computer Science 2024-11-27 Burak Suyunu , Enes Taylan , Arzucan Özgür

The dependence with text length of the statistical properties of word occurrences has long been considered a severe limitation quantitative linguistics. We propose a simple scaling form for the distribution of absolute word frequencies…

Physics and Society · Physics 2015-06-15 Francesc Font-Clos , Gemma Boleda , Álvaro Corral

Natural languages are full of rules and exceptions. One of the most famous quantitative rules is Zipf's law which states that the frequency of occurrence of a word is approximately inversely proportional to its rank. Though this `law' of…

Computation and Language · Computer Science 2015-05-27 Jake Ryland Williams , James P. Bagrow , Christopher M. Danforth , Peter Sheridan Dodds

Recent studies have investigated siamese network architectures for learning invariant speech representations using same-different side information at the word level. Here we investigate systematically an often ignored component of siamese…

Computation and Language · Computer Science 2018-08-24 Rachid Riad , Corentin Dancette , Julien Karadayi , Neil Zeghidour , Thomas Schatz , Emmanuel Dupoux

We show how Zipf's Law can be used to scale up language modeling (LM) to take advantage of more training data and more GPUs. LM plays a key role in many important natural language applications such as speech recognition and machine…

Computation and Language · Computer Science 2018-10-25 Mostofa Patwary , Milind Chabbi , Heewoo Jun , Jiaji Huang , Gregory Diamos , Kenneth Church

The frequencies at which individual words occur across languages follow power law distributions, a pattern of findings known as Zipf's law. A vast literature argues over whether this serves to optimize the efficiency of human communication,…

Computation and Language · Computer Science 2020-01-16 Michael Ramscar

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

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

We study a deliberately simple, fully non-linguistic model of text: a sequence of independent draws from a finite alphabet of letters plus a single space symbol. A word is defined as a maximal block of non-space symbols. Within this…

Computation and Language · Computer Science 2025-11-25 Vladimir Berman

Tokenization and transfer learning are two critical components in building state of the art time series foundation models for forecasting. In this work, we systematically study the effect of tokenizer design, specifically scaling and…

Machine Learning · Computer Science 2025-11-18 Alexis Roger , Gwen Legate , Kashif Rasul , Yuriy Nevmyvaka , Irina Rish

Tokenization - the practice of converting strings of characters from an alphabet into sequences of tokens over a vocabulary - is a critical step in the NLP pipeline. The use of token representations is widely credited with increased model…

Computation and Language · Computer Science 2025-04-04 Juan Luis Gastaldi , John Terilla , Luca Malagutti , Brian DuSell , Tim Vieira , Ryan Cotterell

Zipf's law on word frequency is observed in English, French, Spanish, Italian, and so on, yet it does not hold for Chinese, Japanese or Korean characters. A model for writing process is proposed to explain the above difference, which takes…

Data Analysis, Statistics and Probability · Physics 2013-05-03 Linyuan Lu , Zi-Ke Zhang , Tao Zhou
‹ Prev 1 2 3 10 Next ›