mlsanityData to 14 Oct 2026

MUSS: Multilevel Subset Selection for Relevance and Diversity

arXiv 2503.11126 ↗First posted 14 Mar 2025cs.LGcited by 2

On the map: Retrieval-augmented generation → RAG pipeline design and evaluation

Abstract (arXiv)

The problem of relevant and diverse subset selection has a wide range of applications, including recommender systems and retrieval-augmented generation (RAG). For example, in recommender systems, one is interested in selecting relevant items, while providing a diversified recommendation. Constrained subset selection problem is NP-hard, and popular approaches such as Maximum Marginal Relevance (MMR) are based on greedy selection. Many real-world applications involve large data, but the original MMR work did not consider distributed selection. This limitation was later addressed by a method called DGDS which allows for a distributed setting using random data partitioning. Here, we exploit structure in the data to further improve both scalability and performance on the target application. We propose MUSS, a novel method that uses a multilevel approach to relevant and diverse selection. In a recommender system application, our method can not only improve the performance up to 4 percent points in precision, but is also 20 to 80 times faster. Our method is also capable of outperforming baselines on RAG-based question answering accuracy. We present a novel theoretical approach for analyzing this type of problems, and show that our method achieves a constant factor approximation of the optimal objective. Moreover, our analysis also resulted in a × 2 tighter bound for DGDS compared to previously known bound. Our code is publicly available at https://github.com/amazon-science/muss.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in RAG pipeline design and evaluation, one of 15 directions in Retrieval-augmented generation. Builds on 25 of its 49 references that are in the map. First posted on arXiv 14 Mar 2025.

Builds on

25 of 49 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Better RAG using Relevant Information Gain16 Jul

Other references · 24

  1. 2018Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answeringcited by 2,164
  2. 2020Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problemscited by 1,920
  3. 2020SummEval: Re-evaluating Summarization Evaluationcited by 674
  4. 2023Recommender Systems with Generative Retrievalcited by 198
  5. 2020Effective Diversity in Population Based Reinforcement Learningcited by 119
  6. 2019Diverse Trajectory Forecasting with Determinantal Point Processescited by 80
  7. 2021Differentiable Quality Diversitycited by 72
  8. 2023D2 Pruning: Message Passing for Balancing Diversity and Difficulty in Data Pruningcited by 69
  9. 2024A Review of Modern Recommender Systems Using Generative Models (Gen-RecSys)cited by 58
  10. 2024Clustering and Ranking: Diversity-preserved Instruction Selection through Expert-aligned Quality Estimationcited by 53
  11. 2018Diversity in Machine Learningcited by 48
  12. 2020Optimal Transport Kernels for Sequential and Parallel Neural Architecture Searchcited by 29
12 more
  1. 2018GDPP: Learning Diverse Generations Using Determinantal Point Processcited by 26
  2. 2024Automated Evaluation of Retrieval-Augmented Language Models with Task-Specific Exam Generationcited by 25
  3. 2014Distributed Submodular Maximizationcited by 20
  4. 2019Exact sampling of determinantal point processes with sublinear time preprocessingcited by 19
  5. 2024Enhancing Recommendation Diversity by Re-ranking with Large Language Modelscited by 19
  6. 2015The Power of Randomization: Distributed Submodular Maximization on Massive Datasetscited by 13
  7. 2015Efficient Sampling for k-Determinantal Point Processescited by 11
  8. 2024Relevance meets Diversity: A User-Centric Framework for Knowledge Exploration through Recommendationscited by 11
  9. 2016Exact Sampling from Determinantal Point Processescited by 5
  10. 2022kNN-Embed: Locally Smoothed Embedding Mixtures For Multi-interest Candidate Retrievalcited by 4
  11. 2022Tackling Neural Architecture Search With Quality Diversity Optimizationcited by 3
  12. 2019Distributed Maximization of Submodular plus Diversity Functions for Multi-label Feature Selection on Huge Datasetscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Principled and Scalable Diversity-Aware Retrieval via Cardinality-Constrained Binary Quadratic Programmingcited by 2
  2. 2026WARP: Wasserstein-Aligned RAG for Population Opinionscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024SMART-RAG: Selection using Determinantal Matrices for Augmented Retrieval
  2. 2026Retriever Portfolios: A Principled Approach to Adaptive RAG
  3. 2026LURE-RAG: Lightweight Utility-driven Reranking for Efficient RAG
  4. 2024Know Your RAG: Dataset Taxonomy and Generation Strategies for Evaluating RAG Systems
  5. 2026DF-RAG: Query-Aware Diversity for Retrieval-Augmented Generation
  6. 2026Welfarist Formulations for Diverse Similarity Search
  7. 2025Relevance Isn't All You Need: Scaling RAG Systems With Inference-Time Compute Via Multi-Criteria Reranking
  8. 2024Stochastic RAG: End-to-End Retrieval-Augmented Generation through Expected Utility Maximization