MSMO-ABSA：多尺度和多目标优化用于跨语言基于方面的情感分析
MSMO-ABSA: Multi-Scale and Multi-Objective Optimization for Cross-Lingual Aspect-Based Sentiment Analysis
AI总结：
本文提出MSMO框架，通过多尺度和多目标优化提升跨语言基于方面的情感分析性能，引入代码切换双语句子增强模型鲁棒性，并结合目标语言知识提升效果。
摘要：
基于方面的情感分析（ABSA）在多语言环境下获得了越来越多的研究关注。然而，大多数研究缺乏更稳健的特征对齐和更精细的方面级对齐。本文提出了一种新的框架MSMO：多尺度和多目标优化用于跨语言ABSA。在多尺度对齐中，我们实现了跨语言句子级和方面级对齐，对齐不同上下文环境中的方面术语特征。具体来说，我们引入了代码切换双语句子到语言判别器和一致性训练模块以增强模型的鲁棒性。在多目标优化中，我们设计了两个优化目标：监督训练和一致性训练，旨在增强跨语言语义对齐。为进一步提高模型性能，我们结合了目标语言的知识。结果表明，MSMO通过在多种语言和模型上实现最先进的性能，显著提升了跨语言ABSA。
英文摘要：
Aspect-based sentiment analysis (ABSA) garnered growing research interest in multilingual contexts in the past. However, the majority of the studies lack more robust feature alignment and finer aspect-level alignment. In this paper, we propose a novel framework, MSMO: Multi-Scale and Multi-Objective optimization for cross-lingual ABSA. During multi-scale alignment, we achieve cross-lingual sentence-level and aspect-level alignment, aligning features of aspect terms in different contextual environments. Specifically, we introduce code-switched bilingual sentences into the language discriminator and consistency training modules to enhance the model's robustness. During multi-objective optimization, we design two optimization objectives: supervised training and consistency training, aiming to enhance cross-lingual semantic alignment. To further improve model performance, we incorporate distilled knowledge of the target language into the model. Results show that MSMO significantly enhances cross-lingual ABSA by achieving state-of-the-art performance across multiple languages and models.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
