Using Local Knowledge Graph Construction to Scale Seq2Seq Models to Multi-Document Inputs Angela Fan, Claire Gardent, Chloé Braud, Antoine Bordes To cite this version: Angela Fan, Claire Gardent, Chloé Braud, Antoine Bordes. Using Local Knowledge Graph Construc- tion 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. 10.18653/v1/D19-1428. hal-02277063 HAL Id: hal-02277063 https://hal.archives-ouvertes.fr/hal-02277063 Submitted on 3 Sep 2019 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Public Domain Using Local Knowledge Graph Construction to Scale Seq2Seq Models to Multi-Document Inputs Angela Fan Claire Gardent Chloe´ Braud Antoine Bordes FAIR / LORIA CNRS / LORIA CNRS / LORIA FAIR angelafan,
[email protected] claire.gardent,
[email protected] 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 meth- ods such as TF-IDF ranking. We propose constructing a local graph structured knowl- edge base for each query, which compresses the web search information and reduces re- dundancy.