Enriching and Controlling Global Semantics for Text Summarization

• URL: http://arxiv.org/abs/2109.10616v1
• Date: Wed, 22 Sep 2021 09:31:50 GMT
• Title: Enriching and Controlling Global Semantics for Text Summarization
• Authors: Thong Nguyen, Anh Tuan Luu, Truc Lu, Tho Quan
• Abstract summary: Transformer-based models have been proven effective in the abstractive summarization task by creating fluent and informative summaries. We introduce a neural topic model empowered with normalizing flow to capture the global semantics of the document, which are then integrated into the summarization model. Our method outperforms state-of-the-art summarization models on five common text summarization datasets.
• Score: 11.037667460077813
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recently, Transformer-based models have been proven effective in the abstractive summarization task by creating fluent and informative summaries. Nevertheless, these models still suffer from the short-range dependency problem, causing them to produce summaries that miss the key points of document. In this paper, we attempt to address this issue by introducing a neural topic model empowered with normalizing flow to capture the global semantics of the document, which are then integrated into the summarization model. In addition, to avoid the overwhelming effect of global semantics on contextualized representation, we introduce a mechanism to control the amount of global semantics supplied to the text generation module. Our method outperforms state-of-the-art summarization models on five common text summarization datasets, namely CNN/DailyMail, XSum, Reddit TIFU, arXiv, and PubMed.

Related papers

• Information-Theoretic Distillation for Reference-less Summarization [67.51150817011617]
  We present a novel framework to distill a powerful summarizer based on the information-theoretic objective for summarization. We start off from Pythia-2.8B as the teacher model, which is not yet capable of summarization. We arrive at a compact but powerful summarizer with only 568M parameters that performs competitively against ChatGPT.
  arXiv  Detail & Related papers  (2024-03-20T17:42:08Z)
• Controllable Topic-Focused Abstractive Summarization [57.8015120583044]
  Controlled abstractive summarization focuses on producing condensed versions of a source article to cover specific aspects. This paper presents a new Transformer-based architecture capable of producing topic-focused summaries.
  arXiv  Detail & Related papers  (2023-11-12T03:51:38Z)
• GRETEL: Graph Contrastive Topic Enhanced Language Model for Long Document Extractive Summarization [22.053942327552583]
  We propose a graph contrastive topic enhanced language model (GRETEL) for capturing global semantic information. GRETEL integrates the hierarchical transformer encoder and the graph contrastive learning to fuse the semantic information from the global document context and the gold summary. Experimental results on both general domain and biomedical datasets demonstrate that our proposed method outperforms SOTA methods.
  arXiv  Detail & Related papers  (2022-08-21T23:09:29Z)
• NEWTS: A Corpus for News Topic-Focused Summarization [9.872518517174498]
  This paper introduces the first topical summarization corpus, based on the well-known CNN/Dailymail dataset. We evaluate a range of existing techniques and analyze the effectiveness of different prompting methods.
  arXiv  Detail & Related papers  (2022-05-31T10:01:38Z)
• Fine-tuning GPT-3 for Russian Text Summarization [77.34726150561087]
  This paper showcases ruGPT3 ability to summarize texts, fine-tuning it on the corpora of Russian news with their corresponding human-generated summaries. We evaluate the resulting texts with a set of metrics, showing that our solution can surpass the state-of-the-art model's performance without additional changes in architecture or loss function.
  arXiv  Detail & Related papers  (2021-08-07T19:01:40Z)
• Dialogue Summarization with Supporting Utterance Flow Modeling and Fact Regularization [58.965859508695225]
  We propose an end-to-end neural model for dialogue summarization with two novel modules. The supporting utterance flow modeling helps to generate a coherent summary by smoothly shifting the focus from the former utterances to the later ones. The fact regularization encourages the generated summary to be factually consistent with the ground-truth summary during model training.
  arXiv  Detail & Related papers  (2021-08-03T03:09:25Z)
• Topic-Guided Abstractive Text Summarization: a Joint Learning Approach [19.623946402970933]
  We introduce a new approach for abstractive text summarization, Topic-Guided Abstractive Summarization. The idea is to incorporate neural topic modeling with a Transformer-based sequence-to-sequence (seq2seq) model in a joint learning framework.
  arXiv  Detail & Related papers  (2020-10-20T14:45:25Z)
• Multi-Fact Correction in Abstractive Text Summarization [98.27031108197944]
  Span-Fact is a suite of two factual correction models that leverages knowledge learned from question answering models to make corrections in system-generated summaries via span selection. Our models employ single or multi-masking strategies to either iteratively or auto-regressively replace entities in order to ensure semantic consistency w.r.t. the source text. Experiments show that our models significantly boost the factual consistency of system-generated summaries without sacrificing summary quality in terms of both automatic metrics and human evaluation.
  arXiv  Detail & Related papers  (2020-10-06T02:51:02Z)
• Extractive Summarization as Text Matching [123.09816729675838]
  This paper creates a paradigm shift with regard to the way we build neural extractive summarization systems. We formulate the extractive summarization task as a semantic text matching problem. We have driven the state-of-the-art extractive result on CNN/DailyMail to a new level (44.41 in ROUGE-1)
  arXiv  Detail & Related papers  (2020-04-19T08:27:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.