Using Local Knowledge Graph Construction to Scale Seq2Seq Models to Multi-Document Inputs

Abstract : Query-based open-domain NLP tasks require information synthesis from long and diverse web results. Current approaches extrac-tively select portions of web text as input to Sequence-to-Sequence models using methods such as TF-IDF ranking. We propose constructing a local graph structured knowledge base for each query, which compresses the web search information and reduces redundancy. We show that by linearizing the graph into a structured input sequence, models can encode the graph representations within a standard Sequence-to-Sequence setting. For two generative tasks with very long text input , long-form question answering and multi-document summarization, feeding graph representations as input can achieve better performance than using retrieved text portions.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02277063
Contributor : Claire Gardent <>
Submitted on : Tuesday, September 3, 2019 - 1:10:09 PM
Last modification on : Friday, September 6, 2019 - 1:22:03 AM

File

output.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02277063, version 1

Collections

Citation

Angela Fan, Claire Gardent, Chloé Braud, Antoine Bordes. Using Local Knowledge Graph Construction to Scale Seq2Seq Models to Multi-Document Inputs. 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing, Nov 2019, Hong Kong, China. ⟨hal-02277063⟩

Share

Metrics

Record views

34

Files downloads

116