English
Related papers

Related papers: Token Masking Improves Transformer-Based Text Clas…

200 papers

Transformer-based document cross-encoder rerankers are a central component of modern information retrieval systems. Despite their success, these models suffer from high computational costs due to processing long query-document sequences at…

Information Retrieval · Computer Science 2026-05-22 Shengyao Zhuang , Zhichao Xu , Ivano Lauriola

Using language models as a remote service entails sending private information to an untrusted provider. In addition, potential eavesdroppers can intercept the messages, thereby exposing the information. In this work, we explore the…

Computation and Language · Computer Science 2024-07-04 Re'em Harel , Yair Elboher , Yuval Pinter

Current language models (LMs) use a fixed, static subword tokenizer. This default choice typically results in degraded efficiency and language capabilities, especially in languages other than English. To address this issue, we challenge the…

Computation and Language · Computer Science 2025-06-12 Darius Feher , Ivan Vulić , Benjamin Minixhofer

This research aims to unravel how large language models (LLMs) iteratively refine token predictions through internal processing. We utilized a logit lens technique to analyze the model's token predictions derived from intermediate…

Computation and Language · Computer Science 2025-06-10 Jaturong Kongmanee

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

Stochastic embedding transitions introduce a probabilistic mechanism for adjusting token representations dynamically during inference, mitigating the constraints imposed through static or deterministic embeddings. A transition framework was…

Computation and Language · Computer Science 2025-08-11 Stefan Whitaker , Colin Sisate , Marcel Windsor , Nikolai Fairweather , Tarquin Goldborough , Oskar Lindenfeld

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

Likelihood training and maximization-based decoding result in dull and repetitive generated texts even when using powerful language models (Holtzman et al., 2019). Adding a loss function for regularization was shown to improve text…

Computation and Language · Computer Science 2021-01-13 Evgeny Lagutin , Daniil Gavrilov , Pavel Kalaidin

Recent advancements in large language models demonstrate that injecting perturbations can substantially enhance extrapolation performance. However, current approaches often rely on discrete perturbations with fixed designs, which limits…

Machine Learning · Statistics 2026-05-14 Zetai Cen , Chenfei Gu , Jin Zhu , Ting Li , Yunxiao Chen , Chengchun Shi

Masking tokens uniformly at random constitutes a common flaw in the pretraining of Masked Language Models (MLMs) such as BERT. We show that such uniform masking allows an MLM to minimize its training objective by latching onto shallow local…

Machine Learning · Computer Science 2020-10-06 Yoav Levine , Barak Lenz , Opher Lieber , Omri Abend , Kevin Leyton-Brown , Moshe Tennenholtz , Yoav Shoham

Masked language modeling has become a widely adopted unsupervised technique to pre-train large language models (LLMs). However, the process of selecting tokens for masking is random, and the percentage of masked tokens is typically fixed…

Computation and Language · Computer Science 2025-05-28 Andrei Jarca , Florinel Alin Croitoru , Radu Tudor Ionescu

While there has been a large body of research attempting to circumvent tokenization for language modeling (Clark et al., 2022; Xue et al., 2022), the current consensus is that it is a necessary initial step for designing state-of-the-art…

Computation and Language · Computer Science 2025-04-11 Nived Rajaraman , Jiantao Jiao , Kannan Ramchandran

Neural machine translation with millions of parameters is vulnerable to unfamiliar inputs. We propose Token Drop to improve generalization and avoid overfitting for the NMT model. Similar to word dropout, whereas we replace dropped token…

Computation and Language · Computer Science 2020-10-22 Huaao Zhang , Shigui Qiu , Xiangyu Duan , Min Zhang

Pretraining large language models (LLMs) with next-token prediction has led to remarkable advances, yet the context-dependent nature of token embeddings in such models results in high intra-class variance and inter-class similarity, thus…

Computation and Language · Computer Science 2026-05-12 Yan Sun , Guoxia Wang , Jinle Zeng , JiaBin Yang , Shuai Li , Li Shen , Dacheng Tao , DianHai Yu , Haifeng Wang

Tokenization is a foundational step in the text process of Large Language Models (LLMs). Texts must be first tokenized into token IDs, which are then input to LLMs. Inefficient tokenization results in long token-ID sequences and will slow…

Computation and Language · Computer Science 2026-05-14 Chong Li , Yingzhuo Deng , Wen Yang , Jiajun Zhang , Chengqing Zong

Tokenisation is the first step in almost all NLP tasks, and state-of-the-art transformer-based language models all use subword tokenisation algorithms to process input text. Existing algorithms have problems, often producing tokenisations…

Computation and Language · Computer Science 2022-10-25 Edward Gow-Smith , Harish Tayyar Madabushi , Carolina Scarton , Aline Villavicencio

Masked language modeling (MLM) pre-training models such as BERT corrupt the input by replacing some tokens with [MASK] and then train a model to reconstruct the original tokens. This is an effective technique which has led to good results…

Computation and Language · Computer Science 2020-06-11 Andy Wagner , Tiyasa Mitra , Mrinal Iyer , Godfrey Da Costa , Marc Tremblay

Self-supervised learning has emerged as a powerful approach for leveraging large-scale unlabeled data to improve model performance in various domains. In this paper, we explore masked self-supervised pre-training for text recognition…

Computer Vision and Pattern Recognition · Computer Science 2025-03-31 Martin Kišš , Michal Hradiš

Structured dropout approaches, such as attention dropout and DropHead, have been investigated to regularize the multi-head attention mechanism in Transformers. In this paper, we propose a new regularization scheme based on token-level…

Computation and Language · Computer Science 2023-10-31 Yangjun Wu , Kebin Fang , Dongxiang Zhang , Han Wang , Hao Zhang , Gang Chen

Pre-trained transformer models shine in many natural language processing tasks and therefore are expected to bear the representation of the input sentence or text meaning. These sentence-level embeddings are also important in…

Computation and Language · Computer Science 2025-02-21 Lukas Stankevičius , Mantas Lukoševičius