CrossMuSim: A Cross-Modal Framework for Music Similarity Retrieval with LLM-Powered Text Description Sourcing and Mining
This work addresses the problem of managing and exploring music content on streaming platforms, offering a novel approach that improves retrieval accuracy, though it is incremental in leveraging existing cross-modal and LLM techniques.
The paper tackles music similarity retrieval by introducing a cross-modal contrastive learning framework that uses text descriptions to guide modeling, overcoming data scarcity with a dual-source acquisition method combining online scraping and LLM prompting. It achieves significant performance improvements over benchmarks, as shown through objective metrics, subjective evaluations, and real-world A/B testing on the Huawei Music platform.
Music similarity retrieval is fundamental for managing and exploring relevant content from large collections in streaming platforms. This paper presents a novel cross-modal contrastive learning framework that leverages the open-ended nature of text descriptions to guide music similarity modeling, addressing the limitations of traditional uni-modal approaches in capturing complex musical relationships. To overcome the scarcity of high-quality text-music paired data, this paper introduces a dual-source data acquisition approach combining online scraping and LLM-based prompting, where carefully designed prompts leverage LLMs' comprehensive music knowledge to generate contextually rich descriptions. Exten1sive experiments demonstrate that the proposed framework achieves significant performance improvements over existing benchmarks through objective metrics, subjective evaluations, and real-world A/B testing on the Huawei Music streaming platform.