A combination of LLM paraphrase augmentation and confidence-weighted fine-tuning is reported to improve aspect category sentiment analysis on four SemEval datasets.
Refining and synthesis: A simple yet effective data augmentation frame- work for cross-domain aspect-based sentiment analysis,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
REJECT 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Semantic-preserved Augmentation with Confidence-weighted Fine-tuning for Aspect Category Sentiment Analysis
A combination of LLM paraphrase augmentation and confidence-weighted fine-tuning is reported to improve aspect category sentiment analysis on four SemEval datasets.