Benefiting from Negative yet Informative Feedback by Contrasting Opposing Sequential Patterns

2025-08-20

AI Summary

Negative feedback signals have been long overlooked in sequential recommendation. Method: This paper proposes a dual-Transformer encoder model that jointly incorporates positive and negative feedback. It constructs adversarial interaction sequences for positive and negative user-item interactions, and introduces a composite loss function combining cross-entropy and contrastive learning to explicitly model both semantic discrepancies and complementary relationships between them—marking the first effort to directly embed informative negative feedback into the sequential recommendation objective. Contributions/Results: (1) A novel adversarial modeling framework improves identification of users’ true interests; (2) Multi-task joint optimization simultaneously enhances recommendation accuracy and suppresses undesirable items. Extensive experiments on multiple benchmark datasets demonstrate significant improvements over state-of-the-art methods: notably higher true positive rates and substantially reduced false positive rates for negative items.

Technology Category

• Search and Optimization: Adversarial Search
• Multiagent Systems: Adversarial Agents
• Machine Learning: Adversarial Learning & Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems

Abstract

We consider the task of learning from both positive and negative feedback in a sequential recommendation scenario, as both types of feedback are often present in user interactions. Meanwhile, conventional sequential learning models usually focus on considering and predicting positive interactions, ignoring that reducing items with negative feedback in recommendations improves user satisfaction with the service. Moreover, the negative feedback can potentially provide a useful signal for more accurate identification of true user interests. In this work, we propose to train two transformer encoders on separate positive and negative interaction sequences. We incorporate both types of feedback into the training objective of the sequential recommender using a composite loss function that includes positive and negative cross-entropy as well as a cleverly crafted contrastive term, that helps better modeling opposing patterns. We demonstrate the effectiveness of this approach in terms of increasing true-positive metrics compared to state-of-the-art sequential recommendation methods while reducing the number of wrongly promoted negative items.

Problem

• Learning from both positive and negative sequential feedback
• Reducing negative item recommendations to improve satisfaction
• Accurately identifying user interests using opposing interaction patterns

Innovation

• Dual transformer encoders for positive and negative sequences
• Composite loss with cross-entropy and contrastive terms
• Contrastive learning to model opposing interaction patterns