Optimizing Deep Transformers for Chinese-Thai Low-Resource Translation

Hao, Wenjie; Xu, Hongfei; Mu, Lingling; Hongying, Zan,

doi:10.1007/978-981-19-7960-6_12

Cited by 5 publications

(2 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Training of the paraphrase model. ParaZh-22M (Hao et al 2022) is a large-scale paraphrase dataset with about 22M sentence pairs. We utilize the NER toolkit (TexSmart) to preprocess the dataset.…”

Section: The Construction Of Positive Samplesmentioning

confidence: 99%

Improving the Robustness of Knowledge-Grounded Dialogue via Contrastive Learning

Wang,

Qu,

Wang

et al. 2024

AAAI

View full text Add to dashboard Cite

Knowledge-grounded dialogue (KGD) learns to generate an informative response based on a given dialogue context and external knowledge (e.g., knowledge graphs; KGs). Recently, the emergence of large language models (LLMs) and pre-training techniques has brought great success to knowledge-grounded dialogue. However, when building KGD systems in real applications, there are various real-world noises that are inevitable to face. For example, the dialogue context might involve perturbations such as misspellings and abbreviations. In addition, KGs typically suffer from incompletion and also might contain erroneous and outdated facts. Such real-world noises pose a challenge to the robustness of KGD systems and hinder their applications in the real world. In this paper, we propose an entity-based contrastive learning framework for improving the robustness of KGD. Specifically, we make use of the entity information in a KGD sample to create both its positive and negative samples which involve semantic-irrelevant and semantic-relevant perturbations, respectively. The contrastive learning framework ensures the KGD model is aware of these two types of perturbations, thus could generate informative responses with the potentially noisy inputs in real applications. Experimental results on three widely-used benchmark datasets show that our method achieves new state-of-the-art performance in terms of automatic evaluation scores, verifying its effectiveness and potentiality. Furthermore, we show that our method is able to generate better responses than comparison models in both the noisy and the few-shot settings.

show abstract

Section: The Construction Of Positive Samplesmentioning

confidence: 99%

Improving the Robustness of Knowledge-Grounded Dialogue via Contrastive Learning

Wang,

Qu,

Wang

et al. 2024

AAAI

View full text Add to dashboard Cite

show abstract

“…By configuring the output language to correspond with the input language, multilingual NMT can generate paraphrases directly, overcoming the first limitation. (2) Paraphrases generated from the paraphraser provide diversity in word selection while preserving the sentence's meaning [12,11]. The meaningpreserving properties of paraphrase models can aid in addressing the second limitation.…”

Section: Introductionmentioning

confidence: 99%

Multilingual Lexical Simplification via Paraphrase Generation

Liu,

Qiang,

et al. 2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

Lexical simplification (LS) methods based on pretrained language models have made remarkable progress, generating potential substitutes for a complex word through analysis of its contextual surroundings. However, these methods require separate pretrained models for different languages and disregard the preservation of sentence meaning. In this paper, we propose a novel multilingual LS method via paraphrase generation, as paraphrases provide diversity in word selection while preserving the sentence’s meaning. We regard paraphrasing as a zero-shot translation task within multilingual neural machine translation that supports hundreds of languages. After feeding the input sentence into the encoder of paraphrase modeling, we generate the substitutes based on a novel decoding strategy that concentrates solely on the lexical variations of the complex word. Experimental results demonstrate that our approach surpasses BERT-based methods and zero-shot GPT3-based method significantly on English, Spanish, and Portuguese.

show abstract

Joint Modeling of Chinese Minority Language Translation Tasks

Guo,

Zan,

2023

2023 International Conference on Asian Language Processing (IALP)

View full text Add to dashboard Cite

Optimizing Deep Transformers for Chinese-Thai Low-Resource Translation

Cited by 5 publications

References 30 publications

Improving the Robustness of Knowledge-Grounded Dialogue via Contrastive Learning

Improving the Robustness of Knowledge-Grounded Dialogue via Contrastive Learning

Multilingual Lexical Simplification via Paraphrase Generation

Joint Modeling of Chinese Minority Language Translation Tasks

Contact Info

Product

Resources

About