Vector Retrieval with Similarity and Diversity: How Hard Is It?
It addresses the challenge of unpredictable retrieval results in knowledge-intensive applications like RAG, though it is incremental as it builds on existing methods like MMR.
This paper tackles the problem of balancing similarity and diversity in dense vector retrieval for semantic queries, proving that the joint optimization is NP-complete and introducing a parameter-free heuristic that outperforms baselines like MMR and k-DPP on scientific QA datasets.
Dense vector retrieval is essential for semantic queries within Natural Language Processing, particularly in knowledge-intensive applications like Retrieval-Augmented Generation (RAG). The ability to retrieve vectors that satisfy both similarity and diversity substantially enhances system performance. Although the Maximal Marginal Relevance (MMR) algorithm is widely used to balance these objectives, its reliance on a manually tuned parameter leads to optimization fluctuations and unpredictable retrieval results. Furthermore, there is a lack of sufficient theoretical analysis on the joint optimization of similarity and diversity in vector retrieval. To address these challenges, this paper introduces a novel approach that characterizes both constraints simultaneously by maximizing the similarity between the query vector and the sum of the selected candidate vectors. We formally define this optimization problem, Vectors Retrieval with Similarity and Diversity (VRSD) , and prove that it is NP-complete, establishing a rigorous theoretical bound on the inherent difficulty of this dual-objective retrieval. Subsequently, we present a parameter-free heuristic algorithm to solve VRSD. Extensive evaluations on multiple scientific QA datasets , incorporating both objective geometric metrics and LLM-simulated subjective assessments, demonstrate that our VRSD heuristic consistently outperforms established baselines, including MMR and Determinantal Point Processes (k-DPP).