تم العثور على ملخصات إطفاء التلقائي في كثير من الأحيان تشويه الحقائق أو اختصاصها في المقال.هذا التناقض بين الملخص والنص الأصلي قد أثر بشكل خطير على قابليته للتطبيق.نقترح نموذج تلخيص الحقائق FASUM لاستخراج ودمج العلاقات الواقعية في عملية توليد الموجز عبر انتباه الرسم البياني.ثم نقوم بتصميم نموذج مصحح واقعي FC لتصحيح الأخطاء الواقعية تلقائيا من الملخصات الناتجة عن الأنظمة الحالية.تظهر النتائج التجريبية أن تلخيص حقائق الحقائق يمكن أن تنتج ملخصات إخراج مع اتساق واقعي أعلى مقارنة بالنظام الحالي، ونموذج التصحيح يحسن الاتساق الواقعي الملخصات المعطاة عن طريق تعديل عدد قليل فقط من الكلمات الرئيسية.
Automatic abstractive summaries are found to often distort or fabricate facts in the article. This inconsistency between summary and original text has seriously impacted its applicability. We propose a fact-aware summarization model FASum to extract and integrate factual relations into the summary generation process via graph attention. We then design a factual corrector model FC to automatically correct factual errors from summaries generated by existing systems. Empirical results show that the fact-aware summarization can produce abstractive summaries with higher factual consistency compared with existing systems, and the correction model improves the factual consistency of given summaries via modifying only a few keywords.
References used
https://aclanthology.org/
Neural abstractive summarization systems have gained significant progress in recent years. However, abstractive summarization often produce inconsisitent statements or false facts. How to automatically generate highly abstract yet factually correct s
Factual inconsistencies existed in the output of abstractive summarization models with original documents are frequently presented. Fact consistency assessment requires the reasoning capability to find subtle clues to identify whether a model-generat
In this paper, we focus on improving the quality of the summary generated by neural abstractive dialogue summarization systems. Even though pre-trained language models generate well-constructed and promising results, it is still challenging to summar
The paper describes a system for automatic summarization in English language of online news data that come from different non-English languages. The system is designed to be used in production environment for media monitoring. Automatic summarization
Large scale pretrained models have demonstrated strong performances on several natural language generation and understanding benchmarks. However, introducing commonsense into them to generate more realistic text remains a challenge. Inspired from pre