FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4220828618> ?p ?o ?g. }

• W4220828618 endingPage "102913" @default.
• W4220828618 startingPage "102913" @default.
• W4220828618 abstract "Abstractive summarization aims to generate a concise summary covering salient content from single or multiple text documents. Many recent abstractive summarization methods are built on the transformer model to capture long-range dependencies in the input text and achieve parallelization. In the transformer encoder, calculating attention weights is a crucial step for encoding input documents. Input documents usually contain some key phrases conveying salient information, and it is important to encode these phrases completely. However, existing transformer-based summarization works did not consider key phrases in input when determining attention weights. Consequently, some of the tokens within key phrases only receive small attention weights, which is not conducive to encoding the semantic information of input documents. In this paper, we introduce some prior knowledge of key phrases into the transformer-based summarization model and guide the model to encode key phrases. For the contextual representation of each token in the key phrase, we assume the tokens within the same key phrase make larger contributions compared with other tokens in the input sequence. Based on this assumption, we propose the Key Phrase Aware Transformer (KPAT), a model with the highlighting mechanism in the encoder to assign greater attention weights for tokens within key phrases. Specifically, we first extract key phrases from the input document and score the phrases’ importance. Then we build the block diagonal highlighting matrix to indicate these phrases’ importance scores and positions. To combine self-attention weights with key phrases’ importance scores, we design two structures of highlighting attention for each head and the multi-head highlighting attention. Experimental results on two datasets (Multi-News and PubMed) from different summarization tasks and domains show that our KPAT model significantly outperforms advanced summarization baselines. We conduct more experiments to analyze the impact of each part of our model on the summarization performance and verify the effectiveness of our proposed highlighting mechanism." @default.
• W4220828618 created "2022-04-03" @default.
• W4220828618 creator A5022081685 @default.
• W4220828618 creator A5047431109 @default.
• W4220828618 creator A5068873789 @default.
• W4220828618 creator A5081781438 @default.
• W4220828618 date "2022-05-01" @default.
• W4220828618 modified "2023-09-25" @default.
• W4220828618 title "Key phrase aware transformer for abstractive summarization" @default.
• W4220828618 cites W1975879668 @default.
• W4220828618 cites W1978394996 @default.
• W4220828618 cites W1986154307 @default.
• W4220828618 cites W2008692466 @default.
• W4220828618 cites W2066636486 @default.
• W4220828618 cites W2095521617 @default.
• W4220828618 cites W2125655045 @default.
• W4220828618 cites W2149819339 @default.
• W4220828618 cites W2167094078 @default.
• W4220828618 cites W2183341477 @default.
• W4220828618 cites W2598569220 @default.
• W4220828618 cites W2606974598 @default.
• W4220828618 cites W2609923048 @default.
• W4220828618 cites W2740811004 @default.
• W4220828618 cites W2889518897 @default.
• W4220828618 cites W2904790185 @default.
• W4220828618 cites W2946794439 @default.
• W4220828618 cites W2950670227 @default.
• W4220828618 cites W2953280096 @default.
• W4220828618 cites W2962946054 @default.
• W4220828618 cites W2962965405 @default.
• W4220828618 cites W2963204221 @default.
• W4220828618 cites W2963212250 @default.
• W4220828618 cites W2964813902 @default.
• W4220828618 cites W2970419734 @default.
• W4220828618 cites W2998559045 @default.
• W4220828618 cites W3034682120 @default.
• W4220828618 cites W3038519182 @default.
• W4220828618 cites W3048792033 @default.
• W4220828618 cites W3101913037 @default.
• W4220828618 cites W3104404235 @default.
• W4220828618 cites W3125565408 @default.
• W4220828618 cites W3127104351 @default.
• W4220828618 doi "https://doi.org/10.1016/j.ipm.2022.102913" @default.
• W4220828618 hasPublicationYear "2022" @default.
• W4220828618 type Work @default.
• W4220828618 citedByCount "4" @default.
• W4220828618 countsByYear W42208286182023 @default.
• W4220828618 crossrefType "journal-article" @default.
• W4220828618 hasAuthorship W4220828618A5022081685 @default.
• W4220828618 hasAuthorship W4220828618A5047431109 @default.
• W4220828618 hasAuthorship W4220828618A5068873789 @default.
• W4220828618 hasAuthorship W4220828618A5081781438 @default.
• W4220828618 hasBestOaLocation W42208286182 @default.
• W4220828618 hasConcept C104317684 @default.
• W4220828618 hasConcept C111919701 @default.
• W4220828618 hasConcept C118505674 @default.
• W4220828618 hasConcept C121332964 @default.
• W4220828618 hasConcept C154945302 @default.
• W4220828618 hasConcept C165801399 @default.
• W4220828618 hasConcept C170858558 @default.
• W4220828618 hasConcept C185592680 @default.
• W4220828618 hasConcept C204321447 @default.
• W4220828618 hasConcept C23123220 @default.
• W4220828618 hasConcept C26517878 @default.
• W4220828618 hasConcept C2776224158 @default.
• W4220828618 hasConcept C2780719617 @default.
• W4220828618 hasConcept C38652104 @default.
• W4220828618 hasConcept C41008148 @default.
• W4220828618 hasConcept C48145219 @default.
• W4220828618 hasConcept C55493867 @default.
• W4220828618 hasConcept C62520636 @default.
• W4220828618 hasConcept C66322947 @default.
• W4220828618 hasConcept C66746571 @default.
• W4220828618 hasConceptScore W4220828618C104317684 @default.
• W4220828618 hasConceptScore W4220828618C111919701 @default.
• W4220828618 hasConceptScore W4220828618C118505674 @default.
• W4220828618 hasConceptScore W4220828618C121332964 @default.
• W4220828618 hasConceptScore W4220828618C154945302 @default.
• W4220828618 hasConceptScore W4220828618C165801399 @default.
• W4220828618 hasConceptScore W4220828618C170858558 @default.
• W4220828618 hasConceptScore W4220828618C185592680 @default.
• W4220828618 hasConceptScore W4220828618C204321447 @default.
• W4220828618 hasConceptScore W4220828618C23123220 @default.
• W4220828618 hasConceptScore W4220828618C26517878 @default.
• W4220828618 hasConceptScore W4220828618C2776224158 @default.
• W4220828618 hasConceptScore W4220828618C2780719617 @default.
• W4220828618 hasConceptScore W4220828618C38652104 @default.
• W4220828618 hasConceptScore W4220828618C41008148 @default.
• W4220828618 hasConceptScore W4220828618C48145219 @default.
• W4220828618 hasConceptScore W4220828618C55493867 @default.
• W4220828618 hasConceptScore W4220828618C62520636 @default.
• W4220828618 hasConceptScore W4220828618C66322947 @default.
• W4220828618 hasConceptScore W4220828618C66746571 @default.
• W4220828618 hasIssue "3" @default.
• W4220828618 hasLocation W42208286181 @default.
• W4220828618 hasLocation W42208286182 @default.
• W4220828618 hasOpenAccess W4220828618 @default.
• W4220828618 hasPrimaryLocation W42208286181 @default.

• next