Fine-tuned LLMs with multi-language cross-lingual training score highest in multilingual ABSA; smaller models benefit from code-switching; two new German datasets are released.
Chengyan Wu, Bolei Ma, Ningyuan Deng, Yanqing He, and Yun Xue
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Zero-Shot to Full-Resource: Cross-lingual Transfer Strategies for Aspect-Based Sentiment Analysis
Fine-tuned LLMs with multi-language cross-lingual training score highest in multilingual ABSA; smaller models benefit from code-switching; two new German datasets are released.