Opinions Are Not Always Positive: Debiasing Opinion Summarization with Model-Specific and Model-Agnostic Methods

Yanyue Zhang, Yilong Lai, Zhenglin Wang, Pengfei Li, Deyu Zhou, Yulan He


Abstract
As in the existing opinion summary data set, more than 70% are positive texts, the current opinion summarization approaches are reluctant to generate the negative opinion summary given the input of negative opinions. To address such sentiment bias, two approaches are proposed through two perspectives: model-specific and model-agnostic. For the model-specific approach, a variational autoencoder is proposed to disentangle the input representation into sentiment-relevant and sentiment-irrelevant components through adversarial loss. Therefore, the sentiment information in the input is kept and employed for the following decoding which avoids interference of content information with emotional signals. To further avoid relying on some specific opinion summarization frameworks, a model-agnostic approach based on counterfactual data augmentation is proposed. A dataset with a more balanced emotional polarity distribution is constructed using a large pre-trained language model based on some pairwise and mini-edited principles. Experimental results show that the sentiment consistency of the generated summaries is significantly improved using the proposed approaches, while their semantics quality is unaffected.
Anthology ID:
2024.lrec-main.1094
Volume:
Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
Month:
May
Year:
2024
Address:
Torino, Italia
Editors:
Nicoletta Calzolari, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, Nianwen Xue
Venues:
LREC | COLING
SIG:
Publisher:
ELRA and ICCL
Note:
Pages:
12496–12513
Language:
URL:
https://aclanthology.org/2024.lrec-main.1094
DOI:
Bibkey:
Cite (ACL):
Yanyue Zhang, Yilong Lai, Zhenglin Wang, Pengfei Li, Deyu Zhou, and Yulan He. 2024. Opinions Are Not Always Positive: Debiasing Opinion Summarization with Model-Specific and Model-Agnostic Methods. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 12496–12513, Torino, Italia. ELRA and ICCL.
Cite (Informal):
Opinions Are Not Always Positive: Debiasing Opinion Summarization with Model-Specific and Model-Agnostic Methods (Zhang et al., LREC-COLING 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.lrec-main.1094.pdf