COHEWL: Classifying and Measuring SemanticCoherence of Short Texts with Language Models

Author:

Junior Osmar de Oliveira Braz1,Fileto Renato2

Affiliation:

1. State Univ. of Santa Catarina (UDESC)

2. Federal University of Santa Catarina (UFSC)

Abstract

Abstract

Traditional text coherence models are unable to detect incoherences caused by word misuse in single-sentence documents, as they focus on sentence ordering and semantic similarity of neighboring sentences.This work investigates methods to classify and measure semantic consistency of words in very short documents. Firstly, we fine-tuned BERT for the tasks of detecting short documents with an incoherent word, and distinguishing original documents from the ones with a word automatically changed by the BERT Masked Language Model (MLM). We also used BERT embeddings to calculate coherence measures.Then we prompted generative Large Language Models (LLMs) to classify and measure semantic coherence.The classifiers based on BERT achieved between \(80%\) and \(87.50%\) accuracy in the task of classifying semantic coherence, depending on the language. They performed even better in the task of distinguishing original documents from the ones with a word changed. However, coherence measures calculated using BERT embeddings did not discriminate well coherent documents from incoherent ones, neither original documents from their respective versions with a word automatically changed.On the other hand, LLaMA, GPT, and Gemini outperformed BERT in the task of semantic coherence classification on our corpus of short questions about data structures, in Portuguese and in English. They also generated semantic coherence measures that discriminate coherent from incoherent documents better than measures based on BERT embeddings.

Publisher

Springer Science and Business Media LLC

Reference58 articles.

1. Aletras, Nikolaos and Stevenson, Mark (2013) Evaluating topic coherence using distributional semantics. Association for Computational Linguistics, Potsdam, Germany, https://aclanthology.org/W13-0102/, 13--22, Proceedings of the 10th international conference on computational semantics (IWCS 2013)

2. Barzilay, Regina and Lee, Lillian (2004) Catching the Drift: Probabilistic Content Models, with Applications to Generation and Summarization. Association for Computational Linguistics, Boston, Massachusetts, USA, https://aclanthology.org/N04-1015, 113--120, may 2 - may 7", Proceedings of the Human Language Technology Conference of the North {A}merican Chapter of the Association for Computational Linguistics: {HLT}-{NAACL} 2004

3. Barzilay, Regina and Lapata, Mirella (2008) Modeling local coherence: An entity-based approach. Computational Linguistics 34(1): 1-34 https://doi.org/10.1162/coli.2008.34.1.1, 0891-2017, MIT Press, 03

4. Bao, Mengjiao and Li, Jianxin and Zhang, Jian and Peng, Hao and Liu, Xudong (2019) Learning Semantic Coherence for Machine Generated Spam Text Detection. IEEE, Budapest, Hungary, 10.1109/IJCNN.2019.8852340, 1--8, 2019 Intl. Joint Conf. on Neural Networks

5. De Beaugrande, Robert-Alain and Dressler, Wolfgang U. (1981) Introduction to Text Linguistics. Longman, London, 1

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3