{"url":"/sota/dependency-parsing-on-penn-treebank","task":{"name":"Dependency Parsing","url":"/task/dependency-parsing","note":null},"dataset":{"name":"Penn Treebank","url":"/dataset/penn-treebank"},"category":"Natural Language Processing","categories":["Natural Language Processing"],"category_note":null,"description":"Dependency parsing is the task of extracting a dependency parse of a sentence that represents its grammatical\nstructure and defines the relationships between \"head\" words and words, which modify those heads.\n\nExample:\n\n```\n     root\n      |\n      | +-------dobj---------+\n      | |                    |\nnsubj | |   +------det-----+ | +-----nmod------+\n+--+  | |   |              | | |               |\n|  |  | |   |      +-nmod-+| | |      +-case-+ |\n+  |  + |   +      +      || + |      +      | |\nI  prefer  the  morning   flight  through  Denver\n```\n\nRelations among the words are illustrated above the sentence with directed, labeled\narcs from heads to dependents (+ indicates the dependent).","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["LAS","UAS","POS"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"LAS":null,"UAS":null,"POS":null}},"counts":{"rows":22,"rows_with_code":19,"rows_with_paper_page":21,"rows_dated":20,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"Label Attention Layer + HPSG + XLNet","metrics":{"LAS":"96.26","POS":"97.3","UAS":"97.42"},"uses_additional_data":false,"paper_date":"2019-11-10","paper":"/paper/rethinking-self-attention-an-interpretable","paper_url":"https://arxiv.org/abs/1911.03875v3","paper_title":"Rethinking Self-Attention: Towards Interpretability in Neural Parsing","code":"https://github.com/KhalilMrini/LAL-Parser","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":2,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":2,"model":"DMPar + XLNet","metrics":{"LAS":"95.92","UAS":"97.30"},"uses_additional_data":false,"paper_date":null,"paper":"/paper/enhancing-structure-aware-encoder-with","paper_url":"https://aclanthology.org/2022.coling-1.483","paper_title":"Enhancing Structure-aware Encoder with Extremely Limited Data for Graph-based Dependency Parsing","code":"https://github.com/synlp/dmpar","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"ACE","metrics":{"LAS":"95.8","UAS":"97.2"},"uses_additional_data":false,"paper_date":"2020-10-10","paper":"/paper/automated-concatenation-of-embeddings-for-1","paper_url":"https://arxiv.org/abs/2010.05006v4","paper_title":"Automated Concatenation of Embeddings for Structured Prediction","code":"https://github.com/Alibaba-NLP/ACE","n_code_links":2,"syntology":null},{"rank_in_archive_order":4,"model":"Deep Biaffine + RoBERTa","metrics":{"LAS":"95.75","UAS":"97.29"},"uses_additional_data":false,"paper_date":"2016-11-06","paper":"/paper/deep-biaffine-attention-for-neural-dependency","paper_url":"http://arxiv.org/abs/1611.01734v3","paper_title":"Deep Biaffine Attention for Neural Dependency Parsing","code":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/examples/dependency_parsing/ddparser","n_code_links":26,"syntology":{"n_ran":3,"n_unverified":16,"n_samples":19,"n_pointer_only_licence":2}},{"rank_in_archive_order":5,"model":"HPSG Parser (Joint) + XLNet","metrics":{"LAS":"95.72","POS":"97.3","UAS":"97.20"},"uses_additional_data":false,"paper_date":"2019-07-05","paper":"/paper/head-driven-phrase-structure-grammar-parsing","paper_url":"https://arxiv.org/abs/1907.02684v4","paper_title":"Head-Driven Phrase Structure Grammar Parsing on Penn Treebank","code":"https://github.com/DoodleJZ/HPSG-Neural-Parser","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":7,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"MFVI","metrics":{"LAS":"95.34","UAS":"96.91"},"uses_additional_data":false,"paper_date":"2020-10-10","paper":"/paper/second-order-neural-dependency-parsing-with","paper_url":"https://arxiv.org/abs/2010.05003v2","paper_title":"Second-Order Neural Dependency Parsing with Message Passing and End-to-End Training","code":"https://github.com/wangxinyu0922/Second_Order_Parsing","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"CVT + Multi-Task","metrics":{"LAS":"95.02","UAS":"96.61"},"uses_additional_data":false,"paper_date":"2018-09-22","paper":"/paper/semi-supervised-sequence-modeling-with-cross","paper_url":"http://arxiv.org/abs/1809.08370v1","paper_title":"Semi-Supervised Sequence Modeling with Cross-View Training","code":"https://github.com/tensorflow/models","n_code_links":2,"syntology":null},{"rank_in_archive_order":8,"model":"RNG Transformer","metrics":{"LAS":"95.01","UAS":"96.66"},"uses_additional_data":false,"paper_date":"2020-03-29","paper":"/paper/recursive-non-autoregressive-graph-to-graph","paper_url":"https://arxiv.org/abs/2003.13118v2","paper_title":"Recursive Non-Autoregressive Graph-to-Graph Transformer for Dependency Parsing with Iterative Refinement","code":"https://github.com/idiap/g2g-transformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"SpanRel","metrics":{"LAS":"94.70","UAS":"96.44"},"uses_additional_data":false,"paper_date":"2019-11-10","paper":"/paper/generalizing-natural-language-analysis-1","paper_url":"https://arxiv.org/abs/1911.03822v2","paper_title":"Generalizing Natural Language Analysis through Span-relation Representations","code":"https://github.com/jzbjyb/SpanRel","n_code_links":3,"syntology":null},{"rank_in_archive_order":10,"model":"CRFPar","metrics":{"LAS":"94.49","UAS":"96.14"},"uses_additional_data":false,"paper_date":"2020-05-03","paper":"/paper/efficient-second-order-treecrf-for-neural","paper_url":"https://arxiv.org/abs/2005.00975v2","paper_title":"Efficient Second-Order TreeCRF for Neural Dependency Parsing","code":"https://github.com/yzhangcs/parser","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"Left-to-Right Pointer Network","metrics":{"LAS":"94.43","UAS":"96.04"},"uses_additional_data":false,"paper_date":"2019-03-20","paper":"/paper/left-to-right-dependency-parsing-with-pointer","paper_url":"http://arxiv.org/abs/1903.08445v1","paper_title":"Left-to-Right Dependency Parsing with Pointer Networks","code":"https://github.com/danifg/Left2Right-Pointer-Parser","n_code_links":2,"syntology":null},{"rank_in_archive_order":12,"model":"Graph-based parser with GNNs","metrics":{"LAS":"94.31","POS":"97.3","UAS":"95.97"},"uses_additional_data":false,"paper_date":"2019-07-01","paper":"/paper/graph-based-dependency-parsing-with-graph","paper_url":"https://aclanthology.org/P19-1237","paper_title":"Graph-based Dependency Parsing with Graph Neural Networks","code":"https://github.com/AntNLP/gnn-dep-parsing","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"Deep Biaffine","metrics":{"LAS":"94.22","UAS":"95.87"},"uses_additional_data":false,"paper_date":"2016-11-06","paper":"/paper/deep-biaffine-attention-for-neural-dependency","paper_url":"http://arxiv.org/abs/1611.01734v3","paper_title":"Deep Biaffine Attention for Neural Dependency Parsing","code":"https://github.com/PaddlePaddle/PaddleNLP/tree/develop/examples/dependency_parsing/ddparser","n_code_links":26,"syntology":{"n_ran":3,"n_unverified":16,"n_samples":19,"n_pointer_only_licence":2}},{"rank_in_archive_order":14,"model":"Stack-Pointer Network","metrics":{"LAS":"94.19","POS":"97.3","UAS":"95.87"},"uses_additional_data":false,"paper_date":"2018-05-03","paper":"/paper/stack-pointer-networks-for-dependency-parsing","paper_url":"http://arxiv.org/abs/1805.01087v1","paper_title":"Stack-Pointer Networks for Dependency Parsing","code":"https://github.com/XuezheMax/NeuroNLP2","n_code_links":3,"syntology":null},{"rank_in_archive_order":15,"model":"jPTDP","metrics":{"LAS":"93.87","POS":"97.97","UAS":"95.51"},"uses_additional_data":false,"paper_date":"2018-07-11","paper":"/paper/an-improved-neural-network-model-for-joint","paper_url":"http://arxiv.org/abs/1807.03955v2","paper_title":"An improved neural network model for joint POS tagging and dependency parsing","code":"https://github.com/datquocnguyen/jPTDP","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"Experiment-bert","metrics":{"LAS":"93.2","UAS":"95.42"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":17,"model":"Andor et al.","metrics":{"LAS":"92.79","POS":"97.44","UAS":"94.61"},"uses_additional_data":false,"paper_date":"2016-03-19","paper":"/paper/globally-normalized-transition-based-neural","paper_url":"http://arxiv.org/abs/1603.06042v2","paper_title":"Globally Normalized Transition-Based Neural Networks","code":"https://github.com/tensorflow/models/tree/master/research/syntaxnet","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"Distilled neural FOG","metrics":{"LAS":"92.06","POS":"97.44","UAS":"94.26"},"uses_additional_data":false,"paper_date":"2016-09-24","paper":"/paper/distilling-an-ensemble-of-greedy-dependency","paper_url":"http://arxiv.org/abs/1609.07561v1","paper_title":"Distilling an Ensemble of Greedy Dependency Parsers into One MST Parser","code":"https://github.com/adhigunasurya/distillation_parser","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"Weiss et al.","metrics":{"LAS":"92.06","POS":"97.3","UAS":"94.01"},"uses_additional_data":false,"paper_date":"2015-06-19","paper":"/paper/structured-training-for-neural-network","paper_url":"http://arxiv.org/abs/1506.06158v1","paper_title":"Structured Training for Neural Network Transition-Based Parsing","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":20,"model":"BIST transition-based parser","metrics":{"LAS":"91.9","POS":"97.44","UAS":"93.99"},"uses_additional_data":false,"paper_date":"2016-03-14","paper":"/paper/simple-and-accurate-dependency-parsing-using","paper_url":"http://arxiv.org/abs/1603.04351v3","paper_title":"Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representations","code":"https://github.com/elikip/bist-parser","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":6,"n_samples":7,"n_pointer_only_licence":0}},{"rank_in_archive_order":21,"model":"Arc-hybrid","metrics":{"LAS":"91.42","POS":"97.3","UAS":"93.56"},"uses_additional_data":false,"paper_date":"2016-03-11","paper":"/paper/training-with-exploration-improves-a-greedy","paper_url":"http://arxiv.org/abs/1603.03793v2","paper_title":"Training with Exploration Improves a Greedy Stack-LSTM Parser","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":22,"model":"BIST graph-based parser","metrics":{"LAS":"91.0","POS":"97.3","UAS":"93.1"},"uses_additional_data":false,"paper_date":"2016-03-14","paper":"/paper/simple-and-accurate-dependency-parsing-using","paper_url":"http://arxiv.org/abs/1603.04351v3","paper_title":"Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representations","code":"https://github.com/elikip/bist-parser","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":6,"n_samples":7,"n_pointer_only_licence":0}}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,885 of the 9,623 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9623,"papers_checked":6885,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":7,"rows_with_any_sample_ran":7,"distinct_papers_with_graph_line":5,"distinct_papers_with_any_sample_ran":5,"samples_over_distinct_papers":{"n_ran":9,"n_unverified":32,"n_samples":41,"n_pointer_only_licence":5,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":13,"n_unverified":54,"n_samples":67,"n_pointer_only_licence":7,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}