TY - GEN
T1 - Cross-lingual multi-level adversarial transfer to enhance low-resource name tagging
AU - Huang, Lifu
AU - Ji, Heng
AU - May, Jonathan
N1 - This research is based upon work supported in part by U.S. DARPA LORELEI Program # HR0011-15-C-0115, and the Office of the Director of National Intelligence (ODNI), Intelligence Advanced Research Projects Activity (IARPA), via contract # FA8650-17-C-9116, and ARL NS-CTA No. W911NF-09-2-0053. The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of DARPA, ODNI, IARPA, or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for governmental purposes notwithstanding any copyright annotation therein.
PY - 2019
Y1 - 2019
N2 - We focus on improving name tagging for low-resource languages using annotations from related languages. Previous studies either directly project annotations from a source language to a target language using cross-lingual representations or use a shared encoder in a multitask network to transfer knowledge. These approaches inevitably introduce noise to the target language annotation due to mismatched source-target sentence structures. To effectively transfer the resources, we develop a new neural architecture that leverages multilevel adversarial transfer: (1) word-level adversarial training, which projects source language words into the same semantic space as those of the target language without using any parallel corpora or bilingual gazetteers, and (2) sentence-level adversarial training, which yields language-agnostic sequential features. Our neural architecture outperforms previous approaches on CoNLL data sets. Moreover, on 10 low-resource languages, our approach achieves up to 16% absolute F-score gain over all high-performing baselines on cross-lingual transfer without using any target-language resources.1.
AB - We focus on improving name tagging for low-resource languages using annotations from related languages. Previous studies either directly project annotations from a source language to a target language using cross-lingual representations or use a shared encoder in a multitask network to transfer knowledge. These approaches inevitably introduce noise to the target language annotation due to mismatched source-target sentence structures. To effectively transfer the resources, we develop a new neural architecture that leverages multilevel adversarial transfer: (1) word-level adversarial training, which projects source language words into the same semantic space as those of the target language without using any parallel corpora or bilingual gazetteers, and (2) sentence-level adversarial training, which yields language-agnostic sequential features. Our neural architecture outperforms previous approaches on CoNLL data sets. Moreover, on 10 low-resource languages, our approach achieves up to 16% absolute F-score gain over all high-performing baselines on cross-lingual transfer without using any target-language resources.1.
UR - https://www.scopus.com/pages/publications/85079794429
UR - https://www.scopus.com/pages/publications/85079794429#tab=citedBy
U2 - 10.18653/v1/N19-1383
DO - 10.18653/v1/N19-1383
M3 - Conference contribution
AN - SCOPUS:85079794429
T3 - NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference
SP - 3823
EP - 3833
BT - Long and Short Papers
PB - Association for Computational Linguistics (ACL)
T2 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2019
Y2 - 2 June 2019 through 7 June 2019
ER -