A Neural Framework for English-Hindi Cross-Lingual Natural Language Inference

Saikh, Tanik; De, Arkadipta; Bandyopadhyay, Dibyanayan; Gain, Baban; Ekbal, Asif

doi:10.1007/978-3-030-63830-6_55

Cited by 1 publication

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [26], the authors investigated the effectiveness of language modeling, data augmentation, translation, and architectural approaches to address the codemixed, conversational, and low-resource dataset. Meanwhile, [6] provided a deep neural framework for cross-lingual textual entailment involving English and Hindi. As there are no large datasets available for this task, the authors created their datasets by translating the premises and hypotheses pairs of Stanford Natural Language Inference (SNLI [18]) dataset into Hindi.…”

Section: Related Workmentioning

confidence: 99%

“…Because of being one of the new difficulties suggested for this NLI problem, it is the input sentence that complicates the task. The lack of a large and diverse volume of datasets for this challenge has become the key restriction of research development in this line [6]. Besides, the linguistic (dis)similarity between the languages affects machine learning models to extract the information appropriately.…”

mentioning

confidence: 99%

“…Besides, the linguistic (dis)similarity between the languages affects machine learning models to extract the information appropriately. To solve this challenge, we consider it as a Cross-lingual natural language inference task [6,7,8].…”

mentioning

confidence: 99%

See 2 more Smart Citations

vnNLI - VLSP 2021: Vietnamese and English-Vietnamese Textual Entailment Based on Pre-trained Multilingual Language Models

Thuy

Thìn²,

Vũ³

et al. 2022

JCSCE

View full text Add to dashboard Cite

Natural Language Inference (NLI) is a high-level semantic task in Natural Language Processing - NLP, and it extends further challenges if it is in the cross-lingual scenario. In recent years, pre-trained multilingual language models (e.g., mBERT ,XLM-R, InfoXLM) have greatly contributed to the success of dealing with these challenges. Based on the motivation behind these achievements, this paper describes our approach based on fine-tuning pretrained multilingual language models (XLM-R, InfoXLM) to tackle the shared task ``Vietnamese and English\-Vietnamese Textual Entailment'' at the 8th International Workshop on Vietnamese Language and Speech Processing (VLSP 2021\footnote{https://vlsp.org.vn/vlsp2021}). We investigate other techniques to improve the performance of our work: Cross-validation, Pseudo-labeling (PL), Learning rate adjustment, and Postagging. All experimental results demonstrated that our approach based on the InfoXLM model achieved competitive results, ranking 2nd for the task evaluation in VLSP 2021 with 0.89 in terms of F1-score on the private test set.

show abstract

Section: Related Workmentioning

confidence: 99%

mentioning

confidence: 99%