Reproducibility Analysis And Enhancements For Multi-aspect Dense Retriever With Aspect Learning

Abstract

Multi-aspect dense retrieval aims to incorporate aspect information (e.g., brand and category) into dual encoders to facilitate relevance matching. As an early and representative multi-aspect dense retriever, MADRAL learns several extra aspect embeddings and fuses the explicit aspects with an implicit aspect "OTHER" for final representation. MADRAL was evaluated on proprietary data and its code was not released, making it challenging to validate its effectiveness on other datasets. We failed to reproduce its effectiveness on the public MA-Amazon data, motivating us to probe the reasons and re-examine its components. We propose several component alternatives for comparisons, including replacing "OTHER" with "CLS" and representing aspects with the first several content tokens. Through extensive experiments, we confirm that learning "OTHER" from scratch in aspect fusion is harmful. In contrast, our proposed variants can greatly enhance the retrieval performance. Our research not only shed

Stats

• citations1
• S2 citations—
• github stars9
• HF likes0
• heat score4.26
• arxiv keybi2024reproducibility

Related papers

• A Multi-granularity-aware Aspect Learning Model For Multi-aspect Dense Retrieval (2023)5.24
• Pre-training With Aspect-content Text Mutual Prediction For Multi-aspect Dense Retrieval (2023)5.24
• Multi-aspect Reviewed-item Retrieval Via LLM Query Decomposition And Aspect Fusion (2024)0.00
• Multi-head RAG: Solving Multi-aspect Problems With Llms (2024)0.00
• How To Train Your DRAGON: Diverse Augmentation Towards Generalizable Dense Retrieval (2023)11.39
• Revela: Dense Retriever Learning Via Language Modeling (2025)0.00
• Optimizing Retrieval For RAG Via Reinforcement Learning (2025)0.00
• Investigating Multi-layer Representations For Dense Passage Retrieval (2025)0.00