Cutting-off Redundant Repeating Generations for Neural Abstractive Summarization

TL;DR

This paper introduces a method to reduce redundant repetitions in neural abstractive summarization by estimating vocabulary frequency bounds and controlling output, leading to improved summarization quality.

Contribution

It proposes a novel approach to mitigate repetition in RNN-based models by joint frequency estimation and output control, enhancing summarization performance.

Findings

Significant improvement over baseline models

Achieved best results on a benchmark dataset

Effective reduction of repetitive generation

Abstract

This paper tackles the reduction of redundant repeating generation that is often observed in RNN-based encoder-decoder models. Our basic idea is to jointly estimate the upper-bound frequency of each target vocabulary in the encoder and control the output words based on the estimation in the decoder. Our method shows significant improvement over a strong RNN-based encoder-decoder baseline and achieved its best results on an abstractive summarization benchmark.