{"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/multilingual-constituency-parsing-with-self","title":"Multilingual Constituency Parsing with Self-Attention and Pre-Training","arxiv_id":"1812.11760","date":"2018-12-31","proceeding":"ACL 2019 7","authors":["Nikita Kitaev","Steven Cao","Dan Klein"],"abstract":"We show that constituency parsing benefits from unsupervised pre-training across a variety of languages and a range of pre-training conditions. We first compare the benefits of no pre-training, fastText, ELMo, and BERT for English and find that BERT outperforms ELMo, in large part due to increased model capacity, whereas ELMo in turn outperforms the non-contextual fastText embeddings. We also find that pre-training is beneficial across all 11 languages tested; however, large model sizes (more than 100 million parameters) make it computationally expensive to train separate models for each language. To address this shortcoming, we show that joint multilingual pre-training and fine-tuning allows sharing all but a small number of parameters between ten languages in the final model. The 10x reduction in model size compared to fine-tuning one model per language causes only a 3.2% relative error increase in aggregate. We further explore the idea of joint fine-tuning and show that it gives low-resource languages a way to benefit from the larger datasets of other languages. Finally, we demonstrate new state-of-the-art results for 11 languages, including English (95.8 F1) and Chinese (91.8 F1).","url_abs":"https://arxiv.org/abs/1812.11760v2","url_pdf":"https://arxiv.org/pdf/1812.11760v2.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":"multilingual-constituency-parsing-with-self","repo_url":"https://github.com/nikitakit/self-attentive-parser","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"multilingual-constituency-parsing-with-self","repo_url":"https://github.com/dpfried/rnng-bert","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"multilingual-constituency-parsing-with-self","repo_url":"https://github.com/ghazalkhalighinejad/approximating-cky","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"multilingual-constituency-parsing-with-self","repo_url":"https://github.com/thomaslu2000/incremental-parsing-representations","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"constituency-parsing","task_name":"Constituency Parsing"},{"task_slug":"unsupervised-pre-training","task_name":"Unsupervised Pre-training"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bert","method_name":"BERT"},{"method_slug":"bilstm","method_name":"BiLSTM"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"elmo","method_name":"ELMo"},{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-linear-decay","method_name":"Linear Warmup With Linear Decay"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"},{"method_slug":"fasttext","method_name":"fastText"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/constituency-parsing-on-ctb5","task":"Constituency Parsing","dataset":"CTB5","model":"Kitaev etal. 2019","rank_in_archive_order":6,"of":9,"metrics":{"F1 score":"91.75"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.11760","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1812.11760"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/dpfried/rnng-bert","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nikitakit/self-attentive-parser","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ghazalkhalighinejad/approximating-cky","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/thomaslu2000/incremental-parsing-representations","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":3,"unverified":25},"by_repo_kind":{"official":{"samples":10,"ran":0,"repositories":1},"listed":{"samples":15,"ran":0,"repositories":2}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":3,"samples":[{"code_sha256_prefix":"4289934e599d98d1","entry":"get_graph","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"4289934e599d98d1"}},{"code_sha256_prefix":"bf01d3fec3465c34","entry":"make_command","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"bf01d3fec3465c34"}},{"code_sha256_prefix":"d73ae4737d0072dc","entry":"parse","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"d73ae4737d0072dc"}},{"code_sha256_prefix":"5304ef0f453a56ad","entry":"UniversalTransformerRandomLayerEncoderWithLayer","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"models/ndr.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/models/ndr.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5304ef0f453a56ad"}},{"code_sha256_prefix":"c6fbc9f36f128a4d","entry":"arabic","repo":"nikitakit/self-attentive-parser","repo_kind":"official","path":"src/transliterate.py","file_url":"https://github.com/nikitakit/self-attentive-parser/blob/HEAD/src/transliterate.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c6fbc9f36f128a4d"}},{"code_sha256_prefix":"1248ec1004d55757","entry":"chartscores_from_2hids","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"models/all_models.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/models/all_models.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1248ec1004d55757"}},{"code_sha256_prefix":"6707b9653e35add5","entry":"chartscores_from_hids","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"models/all_models.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/models/all_models.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6707b9653e35add5"}},{"code_sha256_prefix":"bd029ebdc4c2a411","entry":"collapse_unary_strip_pos","repo":"nikitakit/self-attentive-parser","repo_kind":"official","path":"src/benepar/decode_chart.py","file_url":"https://github.com/nikitakit/self-attentive-parser/blob/HEAD/src/benepar/decode_chart.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"bd029ebdc4c2a411"}},{"code_sha256_prefix":"ad00821962c48d10","entry":"collapse_unlabel_binarize","repo":"thomaslu2000/incremental-parsing-representations","repo_kind":"listed","path":"src/tree_transforms.py","file_url":"https://github.com/thomaslu2000/incremental-parsing-representations/blob/HEAD/src/tree_transforms.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ad00821962c48d10"}},{"code_sha256_prefix":"29605c9c3042ed16","entry":"cub_traverse","repo":"thomaslu2000/incremental-parsing-representations","repo_kind":"listed","path":"src/tree_transforms.py","file_url":"https://github.com/thomaslu2000/incremental-parsing-representations/blob/HEAD/src/tree_transforms.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"29605c9c3042ed16"}},{"code_sha256_prefix":"f800c1f3fd0d012a","entry":"evalb","repo":"nikitakit/self-attentive-parser","repo_kind":"official","path":"src/evaluate.py","file_url":"https://github.com/nikitakit/self-attentive-parser/blob/HEAD/src/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f800c1f3fd0d012a"}},{"code_sha256_prefix":"1e2e209296c87494","entry":"extract_parse","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"my_utils.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/my_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1e2e209296c87494"}},{"code_sha256_prefix":"d568621c7cfd96cc","entry":"format_elapsed","repo":"nikitakit/self-attentive-parser","repo_kind":"official","path":"src/export.py","file_url":"https://github.com/nikitakit/self-attentive-parser/blob/HEAD/src/export.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d568621c7cfd96cc"}},{"code_sha256_prefix":"3f9e76811bf1f815","entry":"get_labeled_spans","repo":"nikitakit/self-attentive-parser","repo_kind":"official","path":"src/benepar/decode_chart.py","file_url":"https://github.com/nikitakit/self-attentive-parser/blob/HEAD/src/benepar/decode_chart.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3f9e76811bf1f815"}},{"code_sha256_prefix":"ec1bd125cee0d5c8","entry":"get_potentials_and_trees","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"get_data/get_pcfg.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/get_data/get_pcfg.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ec1bd125cee0d5c8"}},{"code_sha256_prefix":"aa0d9a3ff9c7e9c0","entry":"get_trees_and_vocs","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"get_data/get_pcfg.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/get_data/get_pcfg.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"aa0d9a3ff9c7e9c0"}},{"code_sha256_prefix":"9c0f1355ed00c9d1","entry":"guess_space_after","repo":"nikitakit/self-attentive-parser","repo_kind":"official","path":"src/benepar/ptb_unescape.py","file_url":"https://github.com/nikitakit/self-attentive-parser/blob/HEAD/src/benepar/ptb_unescape.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9c0f1355ed00c9d1"}},{"code_sha256_prefix":"2da31f9e8d6c6497","entry":"guess_space_after_non_english","repo":"nikitakit/self-attentive-parser","repo_kind":"official","path":"src/benepar/ptb_unescape.py","file_url":"https://github.com/nikitakit/self-attentive-parser/blob/HEAD/src/benepar/ptb_unescape.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2da31f9e8d6c6497"}},{"code_sha256_prefix":"98db1b220dd2ecb7","entry":"hebrew","repo":"nikitakit/self-attentive-parser","repo_kind":"official","path":"src/transliterate.py","file_url":"https://github.com/nikitakit/self-attentive-parser/blob/HEAD/src/transliterate.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"98db1b220dd2ecb7"}},{"code_sha256_prefix":"321467c6959866dd","entry":"inner_score2","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"models/all_models.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/models/all_models.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"321467c6959866dd"}},{"code_sha256_prefix":"782a5dffeb3d6b35","entry":"make_voc","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"get_data/get_pcfg.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/get_data/get_pcfg.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"782a5dffeb3d6b35"}},{"code_sha256_prefix":"d501ca5164d5e1b1","entry":"make_voc","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"get_data/make_data.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/get_data/make_data.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d501ca5164d5e1b1"}},{"code_sha256_prefix":"5fa1c233f4e619bb","entry":"pad_charts","repo":"nikitakit/self-attentive-parser","repo_kind":"official","path":"src/benepar/decode_chart.py","file_url":"https://github.com/nikitakit/self-attentive-parser/blob/HEAD/src/benepar/decode_chart.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5fa1c233f4e619bb"}},{"code_sha256_prefix":"8aedd0d17836b284","entry":"ptb_unescape","repo":"nikitakit/self-attentive-parser","repo_kind":"official","path":"src/benepar/ptb_unescape.py","file_url":"https://github.com/nikitakit/self-attentive-parser/blob/HEAD/src/benepar/ptb_unescape.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8aedd0d17836b284"}},{"code_sha256_prefix":"8ec94d08b3904eba","entry":"random_subspan","repo":"thomaslu2000/incremental-parsing-representations","repo_kind":"listed","path":"src/tree_transforms.py","file_url":"https://github.com/thomaslu2000/incremental-parsing-representations/blob/HEAD/src/tree_transforms.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8ec94d08b3904eba"}},{"code_sha256_prefix":"1189a8b722298cac","entry":"sinusoidal_pos_embedding","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"models/ndr.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/models/ndr.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1189a8b722298cac"}},{"code_sha256_prefix":"801802f2f39d3775","entry":"stupid_extract_parse2","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"my_utils.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/my_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"801802f2f39d3775"}},{"code_sha256_prefix":"a7368615fa47e4fe","entry":"tree_from_span","repo":"ghazalkhalighinejad/approximating-cky","repo_kind":"listed","path":"my_utils.py","file_url":"https://github.com/ghazalkhalighinejad/approximating-cky/blob/HEAD/my_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a7368615fa47e4fe"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}