Abstract:Recommender systems (RS) are pivotal in managing information overload in modern digital services. A key challenge in RS is efficiently processing vast item pools to deliver highly personalized recommendations under strict latency constraints. Multi-stage cascade ranking addresses this by employing computationally efficient retrieval methods to cover diverse user interests, followed by more precise ranking models to refine the results. In the retrieval stage, multi-channel retrieval is often used to generate distinct item subsets from different candidate generators, leveraging the complementary strengths of these methods to maximize coverage. However, forwarding all retrieved items overwhelms downstream rankers, necessitating truncation. Despite advancements in individual retrieval methods, multi-channel fusion, the process of efficiently merging multi-channel retrieval results, remains underexplored. We are the first to identify and systematically investigate multi-channel fusion in the retrieval stage. Current industry practices often rely on heuristic approaches and manual designs, which often lead to suboptimal performance. Moreover, traditional gradient-based methods like SGD are unsuitable for this task due to the non-differentiable nature of the selection process. In this paper, we explore advanced channel fusion strategies by assigning systematically optimized weights to each channel. We utilize black-box optimization techniques, including the Cross Entropy Method and Bayesian Optimization for global weight optimization, alongside policy gradient-based approaches for personalized merging. Our methods enhance both personalization and flexibility, achieving significant performance improvements across multiple datasets and yielding substantial gains in real-world deployments, offering a scalable solution for optimizing multi-channel fusion in retrieval.

Combining Multiple Retrieval Systems Using Combinatorial Fusion Analysis and Rank-Score Characteristic Function

Combination of Multiple Retrieval Systems Using Rank-Score Function and Cognitive Diversity

Improved Combination of Multiple Retrieval Systems Using a Dynamic Combinatorial Fusion Algorithm.

Comparing System Selection Methods for the Combinatorial Fusion of Multiple Retrieval Systems

A Comparative Study on the Combination of Multiple Retrieval Systems

Combining similarity measures in content-based image retrieval guided by mutual information

Fusion of effective retrieval strategies in the same information retrieval system

A Term Co-Occurrence Algorithm and the Effect of Co-Occurrence Terms on Result Ranking for Information Retrieval

Sequential Combination Methods for Data Clustering Analysis

Rank-Aware Graph Fusion With Contextual Dissimilarity Measurement For Image Retrieval

Fusion-in-T5: Unifying Document Ranking Signals for Improved Information Retrieval

Multi-Index Fusion Via Similarity Matrix Pooling for Image Retrieval

From "Identical" to "Similar": Fusing Retrieved Lists Based on Inter-Document Similarities

Learning to Combine Ad-hoc Ranking Functions for Image Retrieval

COS-Mix: Cosine Similarity and Distance Fusion for Improved Information Retrieval

Combining document representations for known-item search

Unleashing the Potential of Multi-Channel Fusion in Retrieval for Personalized Recommendations

Efficient information retrieval based on a combination of vector space and probabilistic models

Resources and Evaluations for Multi-Distribution Dense Information Retrieval

A Meta-Evaluation of C/W/L/A Metrics: System Ranking Similarity, System Ranking Consistency and Discriminative Power

Learning to Retrieve Opinions