Creator-Side Recommender System: Challenges, Designs, and Applications

TL;DR

DualRec introduces a creator-side recommender system that mirrors the standard user-side architecture to optimize creator satisfaction by selecting suitable users for each item. It demonstrates that retrieval and ranking modules from user-side systems can be adapted with modest modifications, while addressing the unique user availability challenge through a dedicated UAC component. Implemented on Kwai, DualRec improves creator experience and item exposure at scale, validating the practicality of balancing creator and user needs in large recommender ecosystems. The work provides a concrete blueprint for deploying creator-focused recommendations and highlights the importance of availability-aware design in long-tail content ecosystems.

Abstract

Users and creators are two crucial components of recommender systems. Typical recommender systems focus on the user side, providing the most suitable items based on each user's request. In such scenarios, a few items receive a majority of exposures, while many items receive very few. This imbalance leads to poorer experiences and decreased activity among the creators receiving less feedback, harming the recommender system in the long term. To this end, we develop a creator-side recommender system, called DualRec, to answer the following question: how to find the most suitable users for each item to enhance the creators' experience? We show that typical user-side recommendation algorithms, such as retrieval and ranking algorithms, can be adapted into the creator-side versions with just a few modifications. This greatly simplifies algorithm design in DualRec. Moreover, we discuss a unique challenge in DualRec: the user availability issue, which is not present in user-side recommender systems. To tackle this issue, we incorporate a user availability calculation (UAC) module to effectively enhance DualRec's performance. DualRec has already been implemented in Kwai, a short video recommendation system with over 100 millions user and over 10 million creators, significantly improving the experience for creators.

Figures (9)

• Figure 1: Impacts of Engagement on Future Post Probability.
• Figure 2: The DualRec Framework.
• Figure 3: Similarity-based retrievals: (a) $u2u2i$; (b) $u2i2i$; (c) $i2i2u$; (d) $i2u2u$.
• Figure 4: Two-tower model retrieval
• Figure 5: The structure of the prediction model.