Fundación Vicomtech, Basque Research and Technology Alliance (BRTA)(维克姆科技基金会、巴斯克研究与技术联盟(BRTA));University of the Basque Country UPV/EHU(巴斯克国家大学UPV/EHU)
Easy Read text is one of the main forms of access to information for people with reading difficulties. One of the key characteristics of this type of text is the requirement to split sentences into smaller grammatical segments, to facilitate reading. Automated segmentation methods could foster the creation of Easy Read content, but their viability has yet to be addressed. In this work, we study novel methods for the task, leveraging masked and generative language models, along with constituent parsing. We conduct comprehensive automatic and human evaluations in three languages, analysing the strengths and weaknesses of the proposed alternatives, under scarce resource limitations. Our results highlight the viability of automated Easy Read text segmentation and remaining deficiencies compared to expert-driven human segmentation.
Automating Reading Comprehension by Generating Question and Answer Pairs
Indian Institute of Technology Bombay(印度理工学院孟买分校);IITB-Monash Research Academy(印度理工学院孟买分校-莫纳什研究学院);Monash University(莫纳什大学)
基于神经网络的方法代表了从文本生成问题的最新技术水平。现有的工作集中于从文本中生成问题,而不涉及答案生成。此外,我们的分析表明,处理罕见的单词和生成最合适的问题,给出一个候选答案仍然是现有方法面临的挑战。我们提出了一个新的两阶段的过程中产生的问题,答案对的文本。对于第一阶段,我们提出了替代方案,用于使用指针网络对句子中的关键答案的跨度进行编码。在我们的第二阶段,我们采用序列到序列模型的问题生成,增强了丰富的语言特征。最后,全局注意和答案编码用于生成与答案最相关的问题。我们的动机和语言分析在我们的框架中的每个组件的作用,并考虑这些组成。这一分析得到了广泛的实验评价的支持。使用标准的评价指标以及人类的评价,我们的实验结果验证了显着改善的质量,我们的框架产生的问题在国家的最先进的。这里提出的技术代表了更自动化的阅读理解评估的又一步。我们还提出了一个活的系统\footnote{Demo of the system is available at \url{https://www.cse.iitb.ac.in/~vishwajeet/autoqg.html}.}来证明我们的方法的有效性。
Easy-to-Read in Germany: A Survey on its Current State and Available Resources
Darmstadt University of Applied Sciences (Hochschule Darmstadt)(达姆施塔特应用技术大学);HiTZ Zentroa-Ixa, Euskal Herriko Unibertsitatea (UPV/EHU)(巴斯克地区大学(UPV/EHU))
本文综述德国 Leichte Sprache 与 Einfache Sprache 的发展现状,并梳理面向 LS 的现有 NLP 工具与资源。
Automating Legal Interpretation with LLMs: Retrieval, Generation, and Evaluation
Wangxuan Institute of Computer Technology, Peking University(北京大学王轩计算机技术研究所);Institute for Artificial Intelligence, Peking University(北京大学人工智能研究所);Peking University Law School(北京大学法学院)