Resolving Linguistic Asymmetry: Forging Symmetric Multilingual Embeddings Through Asymmetric Contrastive and Curriculum Learning

Spremljeno u:
Bibliografski detalji
Izdano u:Symmetry vol. 17, no. 9 (2025), p. 1386-1407
Glavni autor: Meng Lei
Daljnji autori: Li, Yinlin, Wei, Wei, Yang Caipei
Izdano:
MDPI AG
Teme:
Online pristup:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Oznake: Dodaj oznaku
Bez oznaka, Budi prvi tko označuje ovaj zapis!

MARC

LEADER 00000nab a2200000uu 4500
001 3254649201
003 UK-CbPIL
022 |a 2073-8994 
024 7 |a 10.3390/sym17091386  |2 doi 
035 |a 3254649201 
045 2 |b d20250101  |b d20251231 
084 |a 231635  |2 nlm 
100 1 |a Meng Lei  |u College of Information Engineering, Xuchang University, Xuchang 461000, China; mengl@xcu.edu.cn 
245 1 |a Resolving Linguistic Asymmetry: Forging Symmetric Multilingual Embeddings Through Asymmetric Contrastive and Curriculum Learning 
260 |b MDPI AG  |c 2025 
513 |a Journal Article 
520 3 |a The pursuit of universal, symmetric semantic representations within large language models (LLMs) faces a fundamental challenge: the inherent asymmetry of natural languages. Different languages exhibit vast disparities in syntactic structures, lexical choices, and cultural nuances, making the creation of a truly shared, symmetric embedding space a non-trivial task. This paper aims to address this critical problem by introducing a novel framework to forge robust and symmetric multilingual sentence embeddings. Our approach, named DACL (Dynamic Asymmetric Contrastive Learning), is anchored in two powerful asymmetric learning paradigms: Contrastive Learning and Dynamic Curriculum Learning (DCL). We extend Contrastive Learning to the multilingual context, where it asymmetrically treats semantically equivalent sentences from different languages (positive pairs) and sentences with distinct meanings (negative pairs) to enforce semantic symmetry in the target embedding space. To further refine this process, we incorporate Dynamic Curriculum Learning, which introduces a second layer of asymmetry by dynamically scheduling training instances from easy to hard. This dual-asymmetric strategy enables the model to progressively master complex cross-lingual relationships, starting with more obvious semantic equivalences and advancing to subtler ones. Our comprehensive experiments on benchmark cross-lingual tasks, including sentence retrieval and cross-lingual classification (XNLI, PAWS-X, MLDoc, MARC), demonstrate that DACL significantly outperforms a wide range of established baselines. The results validate our dual-asymmetric framework as a highly effective approach for forging robust multilingual embeddings, particularly excelling in tasks involving complex linguistic asymmetries. Ultimately, this work contributes a novel dual-asymmetric learning framework that effectively leverages linguistic asymmetry to achieve robust semantic symmetry across languages. It offers valuable insights for developing more capable, fair, and interpretable multilingual LLMs, emphasizing that deliberately leveraging asymmetry in the learning process is a highly effective strategy. 
653 |a Natural language 
653 |a Language 
653 |a Dictionaries 
653 |a Curricula 
653 |a Equivalence 
653 |a Task complexity 
653 |a Symmetry 
653 |a Lexical choice 
653 |a Forging 
653 |a Connotation 
653 |a Robustness 
653 |a Asymmetry 
653 |a Linguistics 
653 |a Semantics 
653 |a Large language models 
653 |a Syntax 
653 |a Syntactic structures 
653 |a Instructional scaffolding 
653 |a Multilingualism 
653 |a Language modeling 
653 |a Bilingualism 
653 |a Embedding 
653 |a Sentences 
653 |a Experiments 
653 |a Classification 
653 |a Retrieval 
653 |a Learning 
653 |a Frame analysis 
653 |a Languages 
700 1 |a Li, Yinlin  |u State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, No. 95 Zhongguancun East Road, Beijing 100100, China 
700 1 |a Wei, Wei  |u School of Science and Electrical Engineering, Beihang University, Beijing 100190, China 
700 1 |a Yang Caipei  |u Wuhan Second Ship Design and Research Institute, Wuhan 430000, China; ycp03042025@163.com 
773 0 |t Symmetry  |g vol. 17, no. 9 (2025), p. 1386-1407 
786 0 |d ProQuest  |t Engineering Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3254649201/abstract/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch 
856 4 0 |3 Full Text + Graphics  |u https://www.proquest.com/docview/3254649201/fulltextwithgraphics/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3254649201/fulltextPDF/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch