Extracting Summary Knowledge Graphs from Long Documents

Wu, Zeqiu; Koncel-Kedziorski, Rik; Ostendorf, Mari; Hajishirzi, Hannaneh

doi:10.48550/arxiv.2009.09162

Cited by 4 publications

(5 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In general, the KGs can be divided into two main categories depending on their graph construction approaches. Many applications treat the KG as the compact and interpretable intermediate representation of the unstructured data (e.g., the document) (Wu et al, 2020c;. Conceptually, it is almost similar to the IE graph, which we have discussed previously.…”

Section: Question: Who Acted In the Movies Directed By The Director O...mentioning

confidence: 97%

“…The information extraction graph (IE Graph) aims to extract the structural information to represent the high-level information among natural sentences, e.g., text-based documents. These extracted relations that capture relations across distant sentences have been demonstrated helpful in many NLP tasks (Wu et al, 2020c;Vashishth et al, 2018;Gupta et al, 2019). In what follows, we will discuss the technical details on how to construct an IE graph for a given paragraph para Vashishth et al, 2018).…”

Section: Information Extraction Graph Constructionmentioning

confidence: 99%

See 1 more Smart Citation

Graph Neural Networks for Natural Language Processing: A Survey

Wu¹,

Sheng²,

Guo³

et al. 2021

Preprint

View full text Add to dashboard Cite

Deep learning has become the dominant approach in coping with various tasks in Natural Language Processing (NLP). Although text inputs are typically represented as a sequence of tokens, there is a rich variety of NLP problems that can be best expressed with a graph structure. As a result, there is a surge of interests in developing new deep learning techniques on graphs for a large number of NLP tasks. In this survey, we present a comprehensive overview on Graph Neural Networks (GNNs) for Natural Language Processing. We propose a new taxonomy of GNNs for NLP, which systematically organizes existing research of GNNs for NLP along three axes: graph construction, graph representation learning, and graph based encoder-decoder models. We further introduce a large number of NLP applications that are exploiting the power of GNNs and summarize the corresponding benchmark datasets, evaluation metrics, and open-source codes. Finally, we discuss various outstanding challenges for making the full use of GNNs for NLP as well as future research directions. To the best of our knowledge, this is the first comprehensive overview of Graph Neural Networks for Natural Language Processing.

show abstract

Section: Question: Who Acted In the Movies Directed By The Director O...mentioning

confidence: 97%

Section: Information Extraction Graph Constructionmentioning

confidence: 99%

Graph Neural Networks for Natural Language Processing: A Survey

Wu¹,

Sheng²,

Guo³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Knowledge graphs are a good source of latent semantics that can form a skeleton for text generation, stressing the text/token selection of abstractive summarization. This idea has been proven possible through papers such as [23][24][25][26][27], and will be continually built upon in the future.…”

Section: Related Workmentioning

confidence: 99%

Enhancing Abstractive Summarization with Extracted Knowledge Graphs and Multi-Source Transformers

Chen,

Wang,

Yue

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

As the popularity of large language models (LLMs) has risen over the course of the last year, led by GPT-3/4 and especially its productization as ChatGPT, we have witnessed the extensive application of LLMs to text summarization. However, LLMs do not intrinsically have the power to verify the correctness of the information they supply and generate. This research introduces a novel approach to abstractive summarization, aiming to address the limitations of LLMs in that they struggle to understand the truth. The proposed method leverages extracted knowledge graph information and structured semantics as a guide for summarization. Building upon BART, one of the state-of-the-art sequence-to-sequence pre-trained LLMs, multi-source transformer modules are developed as an encoder, which are capable of processing textual and graphical inputs. Decoding is performed based on this enriched encoding to enhance the summary quality. The Wiki-Sum dataset, derived from Wikipedia text dumps, is introduced for evaluation purposes. Comparative experiments with baseline models demonstrate the strengths of the proposed approach in generating informative and relevant summaries. We conclude by presenting our insights into utilizing LLMs with graph external information, which will become a powerful aid towards the goal of factually correct and verified LLMs.

show abstract

“…Chen et al [21] discussed a sentenceselection-based approach to text summarization, whereas Gong et al [22] and Wang et al [23] described how multiple documents could be summarized using topic models. Wu et al [24] suggested a new text-to-graph task for predicting summarized knowledge graphs from long documents, whereas Franciscus et al [25] used a belief graph data model that aggregated words in a semantic order to generate short texts. Zhong et al [26] formulated the extractive summarization task as a semantic text-matching problem, in which a source document and candidate summaries (extracted from the original text) were matched in a semantic space.…”

Section: Related Workmentioning

confidence: 99%

Extractive Summarization of Call Transcripts

Biswas

Iakubovich

2022

IEEE Access

View full text Add to dashboard Cite

Automatic text summarization is one of the most challenging and interesting problems in natural language processing (NLP). Text summarization is the process of extracting the most important information from the text and presenting it concisely in fewer sentences. Call transcript involves textual description of a phone conversation between a customer (caller) and agent(s) (customer representatives). Call transcripts pose unique challenges that are not adequately addressed by most open-source automatic text summarizers, which are developed to summarize continuous texts such as articles and stories. This paper presents an indigenously developed method that combines topic modeling and sentence selection with punctuation restoration in condensing ill-punctuated or un-punctuated call transcripts to produce more readable summaries. This unique combination is what distinguishes the proposed summarizer from other text summarizers. Extensive testing, evaluation and comparisons, with an open-source, state-of-the-art extractive summarizer using three different pre-trained language models, have demonstrated the efficacy of this summarizer for call transcript summarization. The summaries generated by the proposed summarizer were shown to be more compelling and useful based on multiple criteria.

show abstract

Extracting Summary Knowledge Graphs from Long Documents

Cited by 4 publications

References 31 publications

Graph Neural Networks for Natural Language Processing: A Survey

Graph Neural Networks for Natural Language Processing: A Survey

Enhancing Abstractive Summarization with Extracted Knowledge Graphs and Multi-Source Transformers

Extractive Summarization of Call Transcripts

Contact Info

Product

Resources

About