Beyond Positive History: Re-ranking with List-level Hybrid Feedback

TL;DR

This paper introduces RELIFE, a novel re-ranking method that leverages list-level hybrid feedback to better capture user preferences and behavior patterns, significantly improving re-ranking performance.

Contribution

The paper proposes a new re-ranking framework that models list-level hybrid feedback using three modules and contrastive learning, addressing limitations of item-level feedback approaches.

Findings

RELIFE outperforms state-of-the-art re-ranking methods in experiments.

The modules effectively disentangle preferences, learn context-aware preferences, and extract behavior patterns.

Contrastive learning improves the alignment of user behavior patterns across lists.

Abstract

As the last stage of recommender systems, re-ranking generates a re-ordered list that aligns with the user's preference. However, previous works generally focus on item-level positive feedback as history (e.g., only clicked items) and ignore that users provide positive or negative feedback on items in the entire list. This list-level hybrid feedback can reveal users' holistic preferences and reflect users' comparison behavior patterns manifesting within a list. Such patterns could predict user behaviors on candidate lists, thus aiding better re-ranking. Despite appealing benefits, extracting and integrating preferences and behavior patterns from list-level hybrid feedback into re-ranking multiple items remains challenging. To this end, we propose Re-ranking with List-level Hybrid Feedback (dubbed RELIFE). It captures user's preferences and behavior patterns with three modules: a Disentangled Interest Miner to disentangle the user's preferences into interests and disinterests, a Sequential Preference Mixer to learn users' entangled preferences considering the context of feedback, and a Comparison-aware Pattern Extractor to capture user's behavior patterns within each list. Moreover, for better integration of patterns, contrastive learning is adopted to align the behavior patterns of candidate and historical lists. Extensive experiments show that RELIFE significantly outperforms SOTA re-ranking baselines.