Test-Time Code-Switching for Cross-lingual Aspect Sentiment Triplet Extraction

2025-01-24

AI Summary

To address challenges in cross-lingual Aspect-Sentiment-Triplet Extraction (ASTE)—including poor cross-lingual transferability, inaccurate code-switching boundary detection, and weak handling of out-of-vocabulary words—this paper proposes a Test-Time Code-Switching (TT-CSW) framework. During training, a generative model jointly encodes bilingual triplets; during inference, an alignment-driven code-switching mechanism dynamically enhances semantic alignment and context awareness over monolingual inputs. Crucially, TT-CSW decouples language constraints between training and inference for the first time, eliminating reliance on static dictionaries or pre-aligned resources. Evaluated on four multilingual ASTE benchmarks, TT-CSW achieves an average weighted F1-score improvement of 3.7%. Moreover, fine-tuning only a small-scale generative model enables TT-CSW to outperform ChatGPT and GPT-4 by 14.2% and 5.0%, respectively, demonstrating significantly enhanced generalizability and practicality—especially in low-resource settings.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

Aspect Sentiment Triplet Extraction (ASTE) is a thriving research area with impressive outcomes being achieved on high-resource languages. However, the application of cross-lingual transfer to the ASTE task has been relatively unexplored, and current code-switching methods still suffer from term boundary detection issues and out-of-dictionary problems. In this study, we introduce a novel Test-Time Code-SWitching (TT-CSW) framework, which bridges the gap between the bilingual training phase and the monolingual test-time prediction. During training, a generative model is developed based on bilingual code-switched training data and can produce bilingual ASTE triplets for bilingual inputs. In the testing stage, we employ an alignment-based code-switching technique for test-time augmentation. Extensive experiments on cross-lingual ASTE datasets validate the effectiveness of our proposed method. We achieve an average improvement of 3.7% in terms of weighted-averaged F1 in four datasets with different languages. Additionally, we set a benchmark using ChatGPT and GPT-4, and demonstrate that even smaller generative models fine-tuned with our proposed TT-CSW framework surpass ChatGPT and GPT-4 by 14.2% and 5.0% respectively.

Problem

• Aspect-based Sentiment Triple Extraction
• Cross-lingual Information Utilization
• Out-of-Vocabulary Words Handling

Innovation

• TT-CSW
• Bilingual Conversion
• Out-of-Vocabulary Handling