English
Related papers

Related papers: Entropy-Driven Pre-Tokenization for Byte-Pair Enco…

200 papers

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

In this paper, we give an in-depth analysis on the mathematical problem formulations and the probabilistic optimization explorations for some of the key components in Transformer model [33] in the field of generative AI. We explore and…

Machine Learning · Computer Science 2024-10-25 Fulu Li

Tokenization is the foundational step in all large language model (LLM) pipelines, yet the dominant approach Byte Pair Encoding (BPE) and its variants is inherently script agnostic and optimized for English like morphology. For…

Computation and Language · Computer Science 2026-03-09 Prabhu Raja

Locating bugs is challenging but one of the most important activities in software development and maintenance phase because there are no certain rules to identify all types of bugs. Existing automatic bug localization tools use various…

Software Engineering · Computer Science 2018-02-21 Saikat Chakraborty , Yujian Li , Matt Irvine , Ripon Saha , Baishakhi Ray

Video compression systems must support increasing bandwidth and data throughput at low cost and power, and can be limited by entropy coding bottlenecks. Efficiency can be greatly improved by parallelizing coding, which can be done at much…

Image and Video Processing · Electrical Eng. & Systems 2023-12-05 Amir Said , Hoang Le , Farzad Farhadzadeh

Emotion-cause pair extraction (ECPE) aims to extract emotion clauses and corresponding cause clauses, which have recently received growing attention. Previous methods sequentially encode features with a specified order. They first encode…

Computation and Language · Computer Science 2022-12-06 Junlong Liu , Xichen Shang , Qianli Ma

Cross-lingual alignment of word embeddings play an important role in knowledge transfer across languages, for improving machine translation and other multi-lingual applications. Current unsupervised approaches rely on similarities in…

Computation and Language · Computer Science 2020-11-30 Silviu Oprea , Sourav Dutta , Haytham Assem

Tokenisation is an integral part of the current NLP pipeline. Current tokenisation algorithms such as BPE and Unigram are greedy algorithms -- they make locally optimal decisions without considering the resulting vocabulary as a whole. We…

Computation and Language · Computer Science 2026-05-22 Jan Tempus , Philip Whittington , Craig W. Schmidt , Dennis Komm , Tiago Pimentel

We introduce Tokenization with Split Trees (ToaST), a subword tokenization method that directly optimizes compression under a new recursive inference procedure. ToaST greedily splits each pretoken into a full binary tree using precomputed…

Computation and Language · Computer Science 2026-05-28 Craig W. Schmidt , Michael Krumdick , Adam Wiemerslage , Seth Ebner , Varshini Reddy , Yuval Pinter , Chris Tanner

This paper is devoted to the finite-length analysis of turbo decoding over the binary erasure channel (BEC). The performance of iterative belief-propagation (BP) decoding of low-density parity-check (LDPC) codes over the BEC can be…

Information Theory · Computer Science 2011-05-31 Eirik Rosnes , Øyvind Ytrehus

Subword segmenters like BPE operate as a preprocessing step in neural machine translation and other (conditional) language models. They are applied to datasets before training, so translation or text generation quality relies on the quality…

Computation and Language · Computer Science 2023-05-12 Francois Meyer , Jan Buys

The transmission or storage of signals typically involves data compression. The final processing step in compression systems is generally an entropy coding stage, which converts symbols into a bit stream based on their probability…

Information Theory · Computer Science 2026-01-13 Tilo Strutz , Roman Rischke

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

Soft prompts have emerged as a powerful alternative to adapters in parameter-efficient fine-tuning (PEFT), enabling large language models (LLMs) to adapt to downstream tasks without architectural changes or parameter updates. While prior…

Computation and Language · Computer Science 2026-01-14 Beso Mikaberidze , Teimuraz Saghinadze , Simon Ostermann , Philipp Muller

Medical image segmentation is usually regarded as one of the most important intermediate steps in clinical situations and medical imaging research. Thus, accurately assessing the segmentation quality of the automatically generated…

Image and Video Processing · Electrical Eng. & Systems 2022-01-19 Zhenxi Zhang , Chunna Tian , Jie Li , Zhusi Zhong , Zhicheng Jiao , Xinbo Gao

Subword tokenization requires balancing computational efficiency and vocabulary coverage, which often leads to suboptimal performance on languages and scripts not prioritized during training. We propose to augment pretrained language models…

Computation and Language · Computer Science 2025-08-12 Jonas F. Lotz , Hendra Setiawan , Stephan Peitz , Yova Kementchedjhieva

Because of their capacity-approaching performance, graph-based codes have a wide range of applications, including communications and storage. In these codes, unequal error protection (UEP) can offer performance gains with limited rate loss.…

Information Theory · Computer Science 2021-01-25 Beyza Dabak , Ahmed Hareedy , Alexei Ashikhmin , Robert Calderbank

This article investigates the possibility to use the class entropy of the output of a connectionist phoneme recogniser to predict time boundaries between phonetic classes. The rationale is that the value of the entropy should increase in…

Audio and Speech Processing · Electrical Eng. & Systems 2024-01-15 Giampiero Salvi

Over the last few years, machine learning unlocked previously infeasible features for compression, such as providing guarantees for users' privacy or tailoring compression to specific data statistics (e.g., satellite images or audio…

Information Theory · Computer Science 2026-03-25 Gergely Flamich

Tokenization is fundamental to pretrained language models (PLMs). Existing tokenization methods for Chinese PLMs typically treat each character as an indivisible token. However, they ignore the unique feature of the Chinese writing system…

Computation and Language · Computer Science 2023-02-16 Chenglei Si , Zhengyan Zhang , Yingfa Chen , Fanchao Qi , Xiaozhi Wang , Zhiyuan Liu , Yasheng Wang , Qun Liu , Maosong Sun