EUVIMEDEUVIMED entdecken
Crossref

Bidirectional Counterfactual Distillation for Review-Based Recommendation

Sheng Sang, Shujie Li, Shuaiyang Li, Kang Liu, Teng Li, Wei Jia, Dan Guo, Feng Xue

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (18) · 15680-15688

DOI: 10.1609/aaai.v40i18.38598 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Sheng Sang und 7 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Review-based recommendation methods typically integrate multiple behaviors, including interactions, reviews, and ratings, to model user preferences. To effectively extract preference signals from diverse behaviors, some studies train multiple student models to capture distinct behavioral patterns, and leverage online distillation to facilitate collaborative learning among them. However, we argue that these techniques suffer from bias contamination from rating distributions and feature homogenization during cross-behavior knowledge transfer: (1) Rating distribution bias, arising from non-uniform historical ratings, propagates across behaviors through distillation, contaminating the true preference representations of other behaviors. (2) Static distillation strategies often lead to homogenized behavioral features, hindering the learning of behavior-specific preferences. To address these issues, we propose a novel Bidirectional Counterfactual Distillation (BiCoD) framework for review-based recommendation. In BiCoD, we first design an adversarial counterfactual distillation module to suppress the impact of non-uniform rating distributions on distillation, thereby preventing it from contaminating the user's true preference representations across behaviors. Subsequently, we introduce a stage-aware bidirectional distillation strategy to enhance the distinctiveness of behavioral features, facilitating the effective learning of behavior-specific preferences. Extensive experiments on five real-world datasets validate the effectiveness and superiority of the proposed framework.

Zitieren

APA

Sang, S., Li, S., Li, S., Liu, K., Li, T., Jia, W., Guo, D., & Xue, F. (2026). Bidirectional Counterfactual Distillation for Review-Based Recommendation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(18), 15680-15688. https://doi.org/10.1609/aaai.v40i18.38598

Vancouver

Sang S, Li S, Li S, Liu K, Li T, Jia W, et al. Bidirectional Counterfactual Distillation for Review-Based Recommendation. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(18):15680-15688. doi:10.1609/aaai.v40i18.38598

BibTeX
@article{Sang2026,
  author = {Sang, Sheng and Li, Shujie and Li, Shuaiyang and Liu, Kang and Li, Teng and Jia, Wei and Guo, Dan and Xue, Feng},
  title = {Bidirectional Counterfactual Distillation for Review-Based Recommendation},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {18},
  pages = {15680--15688},
  doi = {10.1609/aaai.v40i18.38598}
}
Kostenlos starten →