# Finding the Right Balance: Relevance and Diversity in LLM Retrieval

arXiv: https://arxiv.org/abs/2610.09412 · HTML: https://mlsanity.com/paper/2610.09412/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation.
- Builds on 22 of its 87 references that are in the map.
- First posted on arXiv 7 Oct 2026.

## Placed in

- Retrieval-augmented generation / [RAG robustness and adaptive retrieval](https://mlsanity.com/rag/rag-robustness-and-adaptive-retrieval/)

## Builds on (most cited first, 22 of 22)

- [From Local to Global: A Graph RAG Approach to Query-Focused Summarization](https://mlsanity.com/paper/2404.16130/) (2024)
- [Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models](https://mlsanity.com/paper/2506.05176/) (2025)
- [Passage Re-ranking with BERT](https://mlsanity.com/paper/1901.04085/) (2019)
- [Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG](https://mlsanity.com/paper/2501.09136/) (2025)
- [Corrective Retrieval Augmented Generation](https://mlsanity.com/paper/2401.15884/) (2024)
- [Retrieval-Augmented Generation: A Comprehensive Survey of Architectures, Enhancements, and Robustness Frontiers](https://mlsanity.com/paper/2506.00054/) (2025)
- [Vendi-RAG: Adaptively Trading-Off Diversity And Quality Significantly Improves Retrieval Augmented Generation With LLMs](https://mlsanity.com/paper/2502.11228/) (2025)
- [ARAGOG: Advanced RAG Output Grading](https://mlsanity.com/paper/2404.01037/) (2024)
- [Better RAG using Relevant Information Gain](https://mlsanity.com/paper/2407.12101/) (2024)
- [Diversity Enhances an LLM's Performance in RAG and Long-context Task](https://mlsanity.com/paper/2502.09017/) (2025)
- [AdaGReS:Adaptive Greedy Context Selection via Redundancy-Aware Scoring for Token-Budgeted RAG](https://mlsanity.com/paper/2512.25052/) (2025)
- [SMART-RAG: Selection using Determinantal Matrices for Augmented Retrieval](https://mlsanity.com/paper/2409.13992/) (2024)
- [From Ranking to Selection: A Simple but Efficient Dynamic Passage Selector for Retrieval Augmented Generation](https://mlsanity.com/paper/2508.09497/) (2025)
- [Cluster-based Adaptive Retrieval: Dynamic Context Selection for RAG Applications](https://mlsanity.com/paper/2511.14769/) (2025)
- [Byte-Exact Deduplication in Retrieval-Augmented Generation: A Three-Regime Empirical Analysis Across Public Benchmarks](https://mlsanity.com/paper/2605.09611/) (2026)
- [From BM25 to Corrective RAG: Benchmarking Retrieval Strategies for Text-and-Table Documents](https://mlsanity.com/paper/2604.01733/) (2026)
- [Principled and Scalable Diversity-Aware Retrieval via Cardinality-Constrained Binary Quadratic Programming](https://mlsanity.com/paper/2604.02554/) (2026)
- [Scaling DPPs for RAG: Density Meets Diversity](https://mlsanity.com/paper/2604.03240/) (2026)
- [DIVERGE: Diversity-Enhanced RAG for Open-Ended Information Seeking](https://mlsanity.com/paper/2602.00238/) (2026)
- [Benchmarking Retrieval Strategies for Biomedical Retrieval-Augmented Generation: A Controlled Empirical Study](https://mlsanity.com/paper/2605.02520/) (2026)
- [Retriever Portfolios: A Principled Approach to Adaptive RAG](https://mlsanity.com/paper/2605.31176/) (2026)
- [Covering the Unseen: Information Demand Coverage Optimization for Retrieval-Augmented Generation](https://mlsanity.com/paper/2606.29328/) (2026)

## Abstract (arXiv)

Retrieval diversification is widely available in retrieval-augmented generation (RAG) frameworks, yet prior studies disagree on whether it improves retrieval and answer quality. We show that its effectiveness varies primarily with candidate-pool redundancy, in a pattern consistent with the number of distinct evidence pieces a query requires. Using controlled near-duplicate injection and production-style overlapping chunking, we find that diversification harms relevance, evidence coverage and answer quality on clean pools, but becomes beneficial on multi-evidence tasks when redundancy causes nearest-neighbor retrieval to select repeated passages. We therefore introduce a query-adaptive rule that diversifies only when the effective number of distinct documents in the nearest-neighbor top-k selection falls below the query's evidence requirement. Computed from existing embeddings, the rule captures most of the achievable gain, transfers across datasets and encoders and automatically reduces to nearest-neighbor retrieval for single-evidence queries. We also introduce RNG-Score, a geometric reranker with an exact nearest-neighbor fallback whose margin indicates duplicate structure. Overall, we conclude that diversification should be used selectively, based on observable redundancy and evidence requirements.