{"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/on-difficulties-of-cross-lingual-transfer","title":"On Difficulties of Cross-Lingual Transfer with Order Differences: A Case Study on Dependency Parsing","arxiv_id":"1811.00570","date":"2018-11-01","proceeding":"NAACL 2019 6","authors":["Wasi Uddin Ahmad","Zhisong Zhang","Xuezhe Ma","Eduard Hovy","Kai-Wei Chang","Nanyun Peng"],"abstract":"Different languages might have different word orders. In this paper, we\ninvestigate cross-lingual transfer and posit that an order-agnostic model will\nperform better when transferring to distant foreign languages. To test our\nhypothesis, we train dependency parsers on an English corpus and evaluate their\ntransfer performance on 30 other languages. Specifically, we compare encoders\nand decoders based on Recurrent Neural Networks (RNNs) and modified\nself-attentive architectures. The former relies on sequential information while\nthe latter is more flexible at modeling word order. Rigorous experiments and\ndetailed analysis shows that RNN-based architectures transfer well to languages\nthat are close to English, while self-attentive models have better overall\ncross-lingual transferability and perform especially well on distant languages.","url_abs":"http://arxiv.org/abs/1811.00570v3","url_pdf":"http://arxiv.org/pdf/1811.00570v3.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":"on-difficulties-of-cross-lingual-transfer","repo_url":"https://github.com/uclanlp/CrossLingualDepParser","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"on-difficulties-of-cross-lingual-transfer","repo_url":"https://github.com/uclanlp/0ShotCLTwithOD","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"cross-lingual-transfer","task_name":"Cross-Lingual Transfer"},{"task_slug":"dependency-parsing","task_name":"Dependency Parsing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1811.00570","atlas_url":"https://app.syntology.ai/?focus=1811.00570","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}