词汇简化(Lexical Simplification, LS)自动将难读的词替换为更易理解的替代词,同时保留句子的原意。LS是文本简化的前置任务,旨在提高文本对各种目标人群的可及性,包括儿童、第二语言学习者、有阅读障碍或识字水平较低的人群。目前已有若干LS数据集。这些LS数据集通常只专注于LS流程中的一个或两个子任务。然而,截至目前,尚未开发出一个覆盖所有LS子任务的单一LS数据集。我们提出了MultiLS,这是第一个允许创建多任务LS数据集的LS框架。我们还提出了MultiLS-PT,这是使用MultiLS框架创建的第一个数据集。我们通过为葡萄牙语执行所有LS子任务来展示MultiLS-PT的潜力,这些子任务包括:(1)词汇复杂度预测(lexical complexity prediction, LCP),(2)替代词生成,以及(3)替代词排序。我们报告了模型性能,涵盖从基于Transformer的模型到更近期的大语言模型(large language models, LLMs)。
英文摘要:
Lexical Simplification (LS) automatically replaces difficult to read words for easier alternatives while preserving a sentence's original meaning. LS is a precursor to Text Simplification with the aim of improving text accessibility to various target demographics, including children, second language learners, individuals with reading disabilities or low literacy. Several datasets exist for LS. These LS datasets specialize on one or two sub-tasks within the LS pipeline. However, as of this moment, no single LS dataset has been developed that covers all LS sub-tasks. We present MultiLS, the first LS framework that allows for the creation of a multi-task LS dataset. We also present MultiLS-PT, the first dataset to be created using the MultiLS framework. We demonstrate the potential of MultiLS-PT by carrying out all LS sub-tasks of (1). lexical complexity prediction (LCP), (2). substitute generation, and (3). substitute ranking for Portuguese. Model performances are reported, ranging from transformer-based models to more recent large language models (LLMs).
Redefining Simplicity: Benchmarking Large Language Models from Lexical to Document Simplification
School of Information and Engineering, Yangzhou University(扬州大学信息工程学院);School of Computer Science and Information Technology, Hefei University of Technology(合肥工业大学计算机与信息学院)
Teaching the Pre-trained Model to Generate Simple Texts for Text Simplification
Wangxuan Institute of Computer Technology, Peking University(北京大学王选计算机研究所);Center for Data Science, Peking University(北京大学数据科学中心);School of Interactive Computing, Georgia Institute of Technology(佐治亚理工学院交互计算学院)
Let's Simplify Step by Step: Guiding LLM Towards Multilingual Unsupervised Proficiency-Controlled Sentence Simplification
Department of Computer Science, School of Information Science and Technology, Guangdong University of Foreign Studies(计算机科学系,信息科学与技术学院,广东外语外贸大学);College of Intelligence and Computing, Tianjin University(智能与计算学院,天津大学);Lionrock AI Lab, China Merchants Research Institute of Advanced Technology(Lionrock AI实验室,中国远洋科技研究院)
Improving Zero-Shot Event Extraction via Sentence Simplification
ACLED和Our World in Data等网站的成功证明了从新闻、社交媒体、博客和论坛等形式的大量文本数据中以结构化格式提取事件的巨大效用。事件提取可以为正在发生的地缘政治危机提供一个窗口,并产生可操作的情报。随着大型预训练语言模型的激增,机器阅读理解(MRC)近年来已成为事件提取的新范式。在这种方法中,事件论元提取被框定为一个提取的问题回答任务。基于MRC的方法的关键优点之一是其执行zero-shot提取的能力。然而,长期依赖性的问题,即,触发词和论元词之间较大的词汇距离以及处理句法复杂的句子的困难困扰着基于MRC的方法。在本文中,我们提出了一种通用的方法来提高性能的MRC为基础的事件提取进行无监督的句子简化指导的MRC模型本身。我们在ICEWS地缘政治事件提取数据集上评估了我们的方法,特别注意“演员”和“目标”参数角色。我们展示了如何这样的上下文简化可以提高基于MRC的事件提取的性能超过5%的演员提取和超过10%的目标提取。
MultiLexNorm++: A Unified Benchmark and a Generative Model for Lexical Normalization for Asian Languages
School of Information Science and Technology (VISTEC)(信息科学与技术学院);University of Information Technology, Ho Chi Minh City(胡志明市信息技术大学);Vietnam National University, Ho Chi Minh City(越南国家大学胡志明市分校);Graduate School of Science and Engineering, Ehime University(Ehime大学研究生院);D3 Center, The University of Osaka(大阪大学D3中心);Chung-Ang University(Chung-Ang大学);IT University of Copenhagen(哥本哈根IT大学)
An In-depth Evaluation of Large Language Models in Sentence Simplification with Error-based Human Assessment
Graduate School of Information Science and Technology, The University of Osaka(信息科学与技术研究生学校,大阪大学);School of Computing, Institute of Science Tokyo(计算学院,科学研究所东京)
Investigating Large Language Models for Complex Word Identification in Multilingual and Multidomain Setups
National University of Science and Technology POLITEHNICA Bucharest(布加勒斯特理工国立大学);National Institute for Research & Development in Informatics - ICI Bucharest(布加勒斯特国家信息学研究与发展研究所);Academy of Romanian Scientists(罗马尼亚科学院);Paris 1 Panthéon-Sorbonne University(巴黎第一大学先贤祠-索邦大学);University of Bucharest(布加勒斯特大学)
Align and Shine: Building High-Quality Sentence-Aligned Corpora for Multilingual Text Simplification
Centre for Translation, Localisation and Interpreting Studies(翻译、本地化与诠释研究中心);School of Languages, Cultures and Societies(语言、文化和社会学院);University of Leeds, UK(利兹大学)
ÚFAL at MultiLexNorm 2021: Improving Multilingual Lexical Normalization by Fine-tuning ByT5
Charles University(查理大学)
我们在W-NUT 2021上展示了多语言词汇规范化(MultiLexNorm)共享任务的获奖作品(van der Goot等人,2021 a),它评估了11种语言的12个社交媒体数据集的词汇规范化系统。我们的解决方案基于预先训练的字节级语言模型ByT 5(Xue et al.,2021 a),我们进一步对合成数据进行预训练,然后对真实的归一化数据进行微调。我们的系统实现了最好的性能,通过一个很大的利润率在内在的评价,也是最好的性能,通过依存句法分析的外在评价。源代码发布于https://github.com/ufal/multilexnorm2021,微调模型发布于https://huggingface.co/ufal。