Filters








3,566 Hits in 4.4 sec

Cross-lingual Semantic Role Labeling with Model Transfer [article]

Hao Fei and Meishan Zhang and Fei Li and Donghong Ji
2020 arXiv   pre-print
Prior studies show that cross-lingual semantic role labeling (SRL) can be achieved by model transfer under the help of universal features.  ...  In this paper, we fill the gap of cross-lingual SRL by proposing an end-to-end SRL model that incorporates a variety of universal features and transfer methods.  ...  Semantic Role Labeling A line of research efforts has been made for semantic role labeling.  ... 
arXiv:2008.10284v1 fatcat:v3zwcyw22faz3ifofq5mrh2gpu

Global Methods for Cross-lingual Semantic Role and Predicate Labelling

Lonneke van der Plas, Marianna Apidianaki, Chenhua Chen
2014 International Conference on Computational Linguistics  
We build two global models, one for predicate labelling and one for role labelling, each tailored to the task at hand.  ...  We address the problem of transferring semantic annotations to new languages using parallel corpora.  ...  The ideas behind their cross-lingual model adaptation resemble the ideas behind our global method for semantic role labelling.  ... 
dblp:conf/coling/PlasAC14 fatcat:3xtkrjlmifchfgc3ojya7y4tti

Scaling up Automatic Cross-Lingual Semantic Role Annotation

Lonneke van der Plas, Paola Merlo, James Henderson
2011 Annual Meeting of the Association for Computational Linguistics  
Previous approaches to cross-lingual transfer of semantic annotations have addressed this problem with encouraging results on a small scale.  ...  Moreover, we improve the quality of the transferred semantic annotations by using a joint syntacticsemantic parser that learns the correlations between syntax and semantics of the target language and smooths  ...  Johansson and Nugues (2006) trained a FrameNet-based semantic role labeller for Swedish on annotations transferred cross-lingually from English parallel data.  ... 
dblp:conf/acl/PlasMH11 fatcat:ednh5b7yf5cdddg6guyqtzbzaa

Cross-Lingual Semantic Role Labeling with High-Quality Translated Training Corpus [article]

Hao Fei and Meishan Zhang and Donghong Ji
2020 arXiv   pre-print
Cross-lingual SRL is one promising way to address the problem, which has achieved great advances with the help of model transferring and annotation projection.  ...  Many efforts of research are devoted to semantic role labeling (SRL) which is crucial for natural language understanding.  ...  Table 2 : 2 Results of cross-lingual transfer from English. Table 3 : 3 Cross-lingual transfer with multiple sources.  ... 
arXiv:2004.06295v2 fatcat:bo2b3c2asjhjpf3xbam74qpejm

Cross-Lingual SRL Based upon Universal Dependencies

Ondřej Pražák, Miloslav Konopík
2017 RANLP 2017 - Recent Advances in Natural Language Processing Meet Deep Learning  
In this paper, we introduce a cross-lingual Semantic Role Labeling (SRL) system with language independent features based upon Universal Dependencies.  ...  Our SRL system is based upon cross-lingual features derived from universal dependency trees and supervised learning that utilizes a maximum entropy classifier.  ...  The results of the cross-lingual experiment measured by the unsupervised metrics (Table 4 -F c 1 row) indicate that the transfer of semantic roles from English to all other languages is possible.  ... 
doi:10.26615/978-954-452-049-6_077 dblp:conf/ranlp/PrazakK17 fatcat:wgwkrvsysjetjldcm4fve2col4

WikiBank: Using Wikidata to Improve Multilingual Frame-Semantic Parsing

Cezar Sas, Meriem Beloucif, Anders Søgaard
2020 International Conference on Language Resources and Evaluation  
We also integrate this form of supervision into an off-the-shelf frame-semantic parser and allow cross-lingual transfer.  ...  for semantic parsers.  ...  Furthermore, we experiment with reducing the full label sets in the corpora to a common, simplified label set, to facilitate cross-lingual transfer.  ... 
dblp:conf/lrec/SasBS20 fatcat:dvaaatbtdfdbffggczeydv3vz4

CLAR: A Cross-Lingual Argument Regularizer for Semantic Role Labeling [article]

Ishan Jindal, Yunyao Li, Siddhartha Brahma, Huaiyu Zhu
2020 arXiv   pre-print
Semantic role labeling (SRL) identifies predicate-argument structure(s) in a given sentence.  ...  To leverage such similarity in annotation space across languages, we propose a method called Cross-Lingual Argument Regularizer (CLAR).  ...  Cross-Lingual Transfer from Target to Source Language: Interestingly, cross-lingual transfer by CLAR also helps improving the performance of languages with abundant training data.  ... 
arXiv:2011.04732v1 fatcat:hrop2bxfhjephm5ojb6zfudphq

Cross-lingual Structure Transfer for Relation and Event Extraction

Ananya Subburathinam, Di Lu, Heng Ji, Jonathan May, Shih-Fu Chang, Avirup Sil, Clare Voss
2019 Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)  
The event argument role labeling model transferred from English to Chinese achieves similar performance as the model trained from Chinese.  ...  Extensive experiments on cross-lingual relation and event transfer among English, Chinese, and Arabic demonstrate that our approach achieves performance comparable to state-of-the-art supervised models  ...  Cross- lingual transfer of semantic role labeling models. In Proceedings of the 51st Annual Meeting of the Asso- ciation for Computational Linguistics.  ... 
doi:10.18653/v1/d19-1030 dblp:conf/emnlp/SubburathinamLJ19 fatcat:3vwt72efyrfn7ojd6pkn64q7w4

Translate and Label! An Encoder-Decoder Approach for Cross-lingual Semantic Role Labeling [article]

Angel Daza, Anette Frank
2019 arXiv   pre-print
We propose a Cross-lingual Encoder-Decoder model that simultaneously translates and generates sentences with Semantic Role Labeling annotations in a resource-poor target language.  ...  We then train our model in a cross-lingual setting to generate new SRL labeled data.  ...  We are grateful to our annotators and tó Eva Mújdricza-Maydt for her assistance with the human evaluation setup.  ... 
arXiv:1908.11326v1 fatcat:uequw67mdnernipvnn7hd3xvv4

Translate and Label! An Encoder-Decoder Approach for Cross-lingual Semantic Role Labeling

Angel Daza, Anette Frank
2019 Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)  
We propose a Cross-lingual Encoder-Decoder model that simultaneously translates and generates sentences with Semantic Role Labeling annotations in a resource-poor target language.  ...  We then train our model in a cross-lingual setting to generate new SRL labeled data.  ...  We are grateful to our annotators and tó Eva Mújdricza-Maydt for her assistance with the human evaluation setup.  ... 
doi:10.18653/v1/d19-1056 dblp:conf/emnlp/DazaF19 fatcat:c2w2hf2ahjayfd6ng4cwblxd4a

Weakly Supervised Cross-lingual Semantic Relation Classification via Knowledge Distillation

Yogarshi Vyas, Marine Carpuat
2019 Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)  
We introduce a cross-lingual relation classifier trained only with English examples and a bilingual dictionary.  ...  On new English-Chinese and English-Hindi test sets, the resulting models largely outperform baselines that more naïvely rely on bilingual embeddings or dictionaries for cross-lingual transfer, and approach  ...  Our approach transfers knowledge from a monolingual teacher model to a cross-lingual student model.  ... 
doi:10.18653/v1/d19-1532 dblp:conf/emnlp/VyasC19 fatcat:fvvosxraxfd6hhideptggeo66i

Cross-lingual Structure Transfer for Zero-resource Event Extraction

Di Lu, Ananya Subburathinam, Heng Ji, Jonathan May, Shih-Fu Chang, Avirup Sil, Clare R. Voss
2020 International Conference on Language Resources and Evaluation  
Most current cross-lingual transfer learning methods for Information Extraction (IE) have been applied to local sequence labeling tasks.  ...  To tackle more complex tasks such as event extraction, we need to transfer graph structures (event trigger linked to multiple arguments with various roles) across languages.  ...  Figure 2 : 2 Figure 2: Overall Framework for Cross-lingual Event Structure Transfer. Figure 3 : 3 Figure 3: Comparison with Learning Curves for Supervised Bi-LSTM models on trigger labeling.  ... 
dblp:conf/lrec/LuSJMCSV20 fatcat:nebmjihr2zhkxk7ullz4c33woq

A Knowledge-Enhanced Adversarial Model for Cross-lingual Structured Sentiment Analysis [article]

Qi Zhang, Jie Zhou, Qin Chen, Qingchun Bai, Jun Xiao, Liang He
2022 arXiv   pre-print
Notably, we propose a Knowledge-Enhanced Adversarial Model () with both implicit distributed and explicit structural knowledge to enhance the cross-lingual transfer.  ...  In this paper, we focus on the cross-lingual structured sentiment analysis task, which aims to transfer the knowledge from the source language to the target one.  ...  [13] applied multi-lingual contextualized word embedding for crosslingual tasks, such as semantic role labeling, dependency parsing and named entity recognition.  ... 
arXiv:2205.15514v1 fatcat:6hqs2ok5qrfmdmnrkusqtaesya

Cross-lingual Model Transfer Using Feature Representation Projection

Mikhail Kozhevnikov, Ivan Titov
2014 Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)  
This approach displays competitive performance on model transfer for semantic role labeling when compared to direct model transfer and annotation projection and suggests interesting directions for further  ...  We propose a novel approach to crosslingual model transfer based on feature representation projection.  ...  by cross-lingual transfer.  ... 
doi:10.3115/v1/p14-2095 dblp:conf/acl/KozhevnikovT14 fatcat:ufmyu2ay65ee5lctwwje3lppqy

Zero-shot Cross-lingual Conversational Semantic Role Labeling [article]

Han Wu, Haochen Tan, Kun Xu, Shuqi Liu, Lianwei Wu, Linqi Song
2022 arXiv   pre-print
While conversational semantic role labeling (CSRL) has shown its usefulness on Chinese conversational tasks, it is still under-explored in non-Chinese languages due to the lack of multilingual CSRL annotations  ...  Our model implicitly learns language-agnostic, conversational structure-aware and semantically rich representations with the hierarchical encoders and elaborately designed pre-training objectives.  ...  To our best knowledge, our work is the first step to cross-lingual CSRL. Conversational semantic role labeling.  ... 
arXiv:2204.04914v1 fatcat:bhniez3z6zgvliankwmit2hwji
« Previous Showing results 1 — 15 out of 3,566 results