Controlling sentiment tendency during abstractive summarization
Desentiment is a method that controls the sentiment tendency of generated summaries. It integrates sentiment proofreading into the training objective to address the absence of sentiment-polarity labels in existing summarization datasets, and it uses prompt learning to steer the model toward a target sentiment polarity while preserving the semantic content of the source text. (Cao & Li, 2025)
References
2025
Information
Desentiment: A New Method to Control Sentimental Tendency During Summary Generation
Desentiment, a method that controls the sentimental tendency during text summarization, enabling generation of summaries with desired sentiment properties.
@article{cao2025desentiment,title={Desentiment: A New Method to Control Sentimental Tendency During Summary Generation},author={Cao, Hongyu and Li, Jinlong},journal={Information},volume={16},number={6},pages={453},year={2025},}