{"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/stack-pointer-networks-for-dependency-parsing","title":"Stack-Pointer Networks for Dependency Parsing","arxiv_id":"1805.01087","date":"2018-05-03","proceeding":"ACL 2018 7","authors":["Xuezhe Ma","Zecong Hu","Jingzhou Liu","Nanyun Peng","Graham Neubig","Eduard Hovy"],"abstract":"We introduce a novel architecture for dependency parsing: \\emph{stack-pointer\nnetworks} (\\textbf{\\textsc{StackPtr}}). Combining pointer\nnetworks~\\citep{vinyals2015pointer} with an internal stack, the proposed model\nfirst reads and encodes the whole sentence, then builds the dependency tree\ntop-down (from root-to-leaf) in a depth-first fashion. The stack tracks the\nstatus of the depth-first search and the pointer networks select one child for\nthe word at the top of the stack at each step. The \\textsc{StackPtr} parser\nbenefits from the information of the whole sentence and all previously derived\nsubtree structures, and removes the left-to-right restriction in classical\ntransition-based parsers. Yet, the number of steps for building any (including\nnon-projective) parse tree is linear in the length of the sentence just as\nother transition-based parsers, yielding an efficient decoding algorithm with\n$O(n^2)$ time complexity. We evaluate our model on 29 treebanks spanning 20\nlanguages and different dependency annotation schemas, and achieve\nstate-of-the-art performance on 21 of them.","url_abs":"http://arxiv.org/abs/1805.01087v1","url_pdf":"http://arxiv.org/pdf/1805.01087v1.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":"stack-pointer-networks-for-dependency-parsing","repo_url":"https://github.com/XuezheMax/NeuroNLP2","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"stack-pointer-networks-for-dependency-parsing","repo_url":"https://github.com/ShannonAI/mrc-for-dependency-parsing","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"stack-pointer-networks-for-dependency-parsing","repo_url":"https://github.com/danifg/SyntacticPointer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"dependency-parsing","task_name":"Dependency Parsing"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/dependency-parsing-on-penn-treebank","task":"Dependency Parsing","dataset":"Penn Treebank","model":"Stack-Pointer Network","rank_in_archive_order":14,"of":22,"metrics":{"LAS":"94.19","POS":"97.3","UAS":"95.87"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.01087","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}