Hallucination Reduction in Long Input Text Summarization

TL;DR

This paper proposes methods to reduce hallucinations in long document summarization by fine-tuning the LED model with data filtering and joint entity-summary generation, improving factual consistency.

Contribution

It introduces a novel approach combining data filtering and JAENS during fine-tuning of LED to minimize hallucinations in long text summaries.

Findings

Data filtering reduces entity-level hallucinations.

Fine-tuned LED improves factual consistency.

Methods enhance summary reliability for scientific documents.

Abstract

Hallucination in text summarization refers to the phenomenon where the model generates information that is not supported by the input source document. Hallucination poses significant obstacles to the accuracy and reliability of the generated summaries. In this paper, we aim to reduce hallucinated outputs or hallucinations in summaries of long-form text documents. We have used the PubMed dataset, which contains long scientific research documents and their abstracts. We have incorporated the techniques of data filtering and joint entity and summary generation (JAENS) in the fine-tuning of the Longformer Encoder-Decoder (LED) model to minimize hallucinations and thereby improve the quality of the generated summary. We have used the following metrics to measure factual consistency at the entity level: precision-source, and F1-target. Our experiments show that the fine-tuned LED model performs well in generating the paper abstract. Data filtering techniques based on some preprocessing steps reduce entity-level hallucinations in the generated summaries in terms of some of the factual consistency metrics.