mFACE: Multilingual Summarization with Factual Consistency Evaluation
This addresses the issue of factual hallucinations in multilingual summarization, which is crucial for real-world applications, though it is incremental as it builds on existing evaluation models.
The paper tackled the problem of factually inconsistent summaries in multilingual abstractive summarization by using a multilingual NLI model for data filtering and controlled generation, achieving gains over strong baselines in 45 languages from the XLSum dataset.
Abstractive summarization has enjoyed renewed interest in recent years, thanks to pre-trained language models and the availability of large-scale datasets. Despite promising results, current models still suffer from generating factually inconsistent summaries, reducing their utility for real-world application. Several recent efforts attempt to address this by devising models that automatically detect factual inconsistencies in machine generated summaries. However, they focus exclusively on English, a language with abundant resources. In this work, we leverage factual consistency evaluation models to improve multilingual summarization. We explore two intuitive approaches to mitigate hallucinations based on the signal provided by a multilingual NLI model, namely data filtering and controlled generation. Experimental results in the 45 languages from the XLSum dataset show gains over strong baselines in both automatic and human evaluation.