4.7 Article

Robust scientific text classification using prompt tuning based on data augmentation with L2 regularization

Related references

Note: Only part of the references are listed.
Article Computer Science, Information Systems

An ensemble model for classifying idioms and literal texts using BERT and RoBERTa

J. Briskilal et al.

Summary: This study proposes a predictive ensemble model for classifying idioms and literals using pre-trained BERT and RoBERTa models, which outperforms baseline models in terms of metrics such as F-score and accuracy, showing a 2% improvement in accuracy.

INFORMATION PROCESSING & MANAGEMENT (2022)

Article Computer Science, Information Systems

Fake news detection via knowledgeable prompt learning

Gongyao Jiang et al.

Summary: Researchers propose a knowledgeable prompt learning (KPL) method for fake news detection. By designing prompt templates and verbal words, and incorporating external knowledge, the KPL method outperforms traditional fine-tuning methods in fake news detection.

INFORMATION PROCESSING & MANAGEMENT (2022)

Article Computer Science, Information Systems

Question classification using limited labelled data

Chindukuri Mallikarjuna et al.

Summary: Question classification is a crucial task in the Question Answering system. Existing methods require a large amount of labeled training data, while this paper proposes an approach that uses data augmentation to generate additional training instances. Experimental results demonstrate its effectiveness on two datasets.

INFORMATION PROCESSING & MANAGEMENT (2022)

Article Computer Science, Theory & Methods

Text Data Augmentation for Deep Learning

Connor Shorten et al.

Summary: In this paper, we discuss how Data Augmentation can aid in the development of Natural Language Processing, covering topics such as the motivation and specific frameworks of Data Augmentation, promising ideas yet to be tested, tools for implementation, and interesting topics around Data Augmentation.

JOURNAL OF BIG DATA (2021)

Proceedings Paper Computer Science, Artificial Intelligence

ImpactCite: An XLNet-based Solution Enabling Qualitative Citation Impact Analysis Utilizing Sentiment and Intent

Dominique Mercier et al.

Summary: This study introduces ImpactCite, an XLNet-based method for citation impact analysis, and presents a clean and reliable dataset for citation sentiment analysis. Evaluation results demonstrate that ImpactCite achieves significant performance improvements in citation intent and sentiment classification, suggesting its potential value in citation impact analysis.

ICAART: PROCEEDINGS OF THE 13TH INTERNATIONAL CONFERENCE ON AGENTS AND ARTIFICIAL INTELLIGENCE - VOL 2 (2021)

Article Computer Science, Artificial Intelligence

How Can We Know What Language Models Know?

Zhengbao Jiang et al.

TRANSACTIONS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (2020)