{"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/scene-graph-parsing-as-dependency-parsing","title":"Scene Graph Parsing as Dependency Parsing","arxiv_id":"1803.09189","date":"2018-03-25","proceeding":"NAACL 2018 6","authors":["Yu-Siang Wang","Chenxi Liu","Xiaohui Zeng","Alan Yuille"],"abstract":"In this paper, we study the problem of parsing structured knowledge graphs\nfrom textual descriptions. In particular, we consider the scene graph\nrepresentation that considers objects together with their attributes and\nrelations: this representation has been proved useful across a variety of\nvision and language applications. We begin by introducing an alternative but\nequivalent edge-centric view of scene graphs that connect to dependency parses.\nTogether with a careful redesign of label and action space, we combine the\ntwo-stage pipeline used in prior work (generic dependency parsing followed by\nsimple post-processing) into one, enabling end-to-end training. The scene\ngraphs generated by our learned neural dependency parser achieve an F-score\nsimilarity of 49.67% to ground truth graphs on our evaluation set, surpassing\nbest previous approaches by 5%. We further demonstrate the effectiveness of our\nlearned parser on image retrieval applications.","url_abs":"http://arxiv.org/abs/1803.09189v1","url_pdf":"http://arxiv.org/pdf/1803.09189v1.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":"scene-graph-parsing-as-dependency-parsing","repo_url":"https://github.com/Yusics/bist-parser","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"scene-graph-parsing-as-dependency-parsing","repo_url":"https://github.com/zhuang-li/factual","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"dependency-parsing","task_name":"Dependency Parsing"},{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"knowledge-graphs","task_name":"Knowledge Graphs"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1803.09189","atlas_url":"https://app.syntology.ai/?focus=1803.09189","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}