Assessment of Transformer-Based Encoder-Decoder Model for Human-Like Summarization

TL;DR

This paper evaluates transformer-based BART models for human-like text summarization, comparing their performance with baseline models and exploring domain adaptation and factual accuracy assessment.

Contribution

It introduces a comprehensive evaluation of BART for abstractive summarization, including human evaluation, domain adaptation, and factual consistency analysis.

Findings

Finetuned BART outperforms baseline in ROUGE and BERTScore metrics.

Human-written summaries are 17% more factually consistent than model-generated ones.

Evaluation metrics like ROUGE and BERTScore are insensitive to factual errors.

Abstract

In recent times, extracting valuable information from large text is making significant progress. Especially in the current era of social media, people expect quick bites of information. Automatic text summarization seeks to tackle this by slimming large texts down into more manageable summaries. This important research area can aid in decision-making by digging out salient content from large text. With the progress in deep learning models, significant work in language models has emerged. The encoder-decoder framework in deep learning has become the central approach for automatic text summarization. This work leverages transformer-based BART model for human-like summarization which is an open-ended problem with many challenges. On training and fine-tuning the encoder-decoder model, it is tested with diverse sample articles and the quality of summaries of diverse samples is assessed based on human evaluation parameters. Further, the finetuned model performance is compared with the baseline pretrained model based on evaluation metrics like ROUGE score and BERTScore. Additionally, domain adaptation of the model is required for improved performance of abstractive summarization of dialogues between interlocutors. On investigating, the above popular evaluation metrics are found to be insensitive to factual errors. Further investigation of the summaries generated by finetuned model is done using the contemporary evaluation metrics of factual consistency like WeCheck and SummaC. Empirical results on BBC News articles highlight that the gold standard summaries written by humans are more factually consistent by 17% than the abstractive summaries generated by finetuned model.