{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/hierarchical-losses-and-new-resources-for","title":"Hierarchical Losses and New Resources for Fine-grained Entity Typing and Linking","arxiv_id":"1807.05127","date":"2018-07-13","proceeding":"ACL 2018 7","authors":["Shikhar Murty*","Patrick Verga*","Luke Vilnis","Irena Radovanovic","Andrew McCallum"],"abstract":"Extraction from raw text to a knowledge base of entities and fine-grained\ntypes is often cast as prediction into a flat set of entity and type labels,\nneglecting the rich hierarchies over types and entities contained in curated\nontologies. Previous attempts to incorporate hierarchical structure have\nyielded little benefit and are restricted to shallow ontologies. This paper\npresents new methods using real and complex bilinear mappings for integrating\nhierarchical information, yielding substantial improvement over flat\npredictions in entity linking and fine-grained entity typing, and achieving new\nstate-of-the-art results for end-to-end models on the benchmark FIGER dataset.\nWe also present two new human-annotated datasets containing wide and deep\nhierarchies which we will release to the community to encourage further\nresearch in this direction: MedMentions, a collection of PubMed abstracts in\nwhich 246k mentions have been mapped to the massive UMLS ontology; and TypeNet,\nwhich aligns Freebase types with the WordNet hierarchy to obtain nearly 2k\nentity types. In experiments on all three datasets we show substantial gains\nfrom hierarchy-aware training.","url_abs":"http://arxiv.org/abs/1807.05127v1","url_pdf":"http://arxiv.org/pdf/1807.05127v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"hierarchical-losses-and-new-resources-for","repo_url":"https://github.com/MurtyShikhar/Hierarchical-Typing","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"hierarchical-losses-and-new-resources-for","repo_url":"https://github.com/chanzuckerberg/MedMentions","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"2k","task_name":"2k"},{"task_slug":"entity-linking","task_name":"Entity Linking"},{"task_slug":"entity-typing","task_name":"Entity Typing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1807.05127","atlas_url":"https://app.syntology.ai/?focus=1807.05127","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}