# Metadata, Structure, or Strategy? A Decomposition of RAG Context Enrichment

arXiv: https://arxiv.org/abs/2606.29645 · HTML: https://mlsanity.com/paper/2606.29645/

## Key facts

- Cited by 1 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 18 of its 26 references that are in the map.
- First posted on arXiv 28 Jun 2026.

## Placed in

- Retrieval-augmented generation / [RAG pipeline design and evaluation](https://mlsanity.com/rag/rag-pipeline-design-and-evaluation/)

## Builds on (most cited first, 18 of 18)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection](https://mlsanity.com/paper/2310.11511/) (2023)
- [FEVER: a large-scale dataset for Fact Extraction and VERification](https://mlsanity.com/paper/1803.05355/) (2018)
- [GAIA: a benchmark for General AI Assistants](https://mlsanity.com/paper/2311.12983/) (2023)
- [Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions](https://mlsanity.com/paper/2212.10509/) (2022)
- [MuSiQue: Multihop Questions via Single-hop Question Composition](https://mlsanity.com/paper/2108.00573/) (2021)
- [Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity](https://mlsanity.com/paper/2403.14403/) (2024)
- [ARES: An Automated Evaluation Framework for Retrieval-Augmented Generation Systems](https://mlsanity.com/paper/2311.09476/) (2023)
- [A Dataset for Answering Time-Sensitive Questions](https://mlsanity.com/paper/2108.06314/) (2021)
- [Memento: Fine-tuning LLM Agents without Fine-tuning LLMs](https://mlsanity.com/paper/2508.16153/) (2025)
- [KG-RAG: Bridging the Gap Between Knowledge and Creativity](https://mlsanity.com/paper/2405.12035/) (2024)
- [PersonaRAG: Enhancing Retrieval-Augmented Generation Systems with User-Centric Agents](https://mlsanity.com/paper/2407.09394/) (2024)
- [T-GRAG: A Dynamic GraphRAG Framework for Resolving Temporal Conflicts and Redundancy in Knowledge Retrieval](https://mlsanity.com/paper/2508.01680/) (2025)
- [OwlerLite: Scope- and Freshness-Aware Web Retrieval for LLM Assistants](https://mlsanity.com/paper/2601.17824/) (2026)
- [UXSim: Towards a Hybrid User Search Simulation](https://mlsanity.com/paper/2602.24241/) (2026)
- [NuggetIndex: Governed Atomic Retrieval for Maintainable RAG](https://mlsanity.com/paper/2604.27306/) (2026)

## Built on by (most cited first, 1 of 1)

- [The Compaction Cliff in Long-Running AI Agent Memory](https://mlsanity.com/paper/2608.22752/) (2026)

## Abstract (arXiv)

Retrieval-augmented generation (RAG) systems increasingly enrich retrieved passages by attaching quality metadata, structuring them into explicit records, and adopting multi-hop retrieval strategies that accumulate evidence across steps. These changes assume that richer context yields better answers, yet existing evaluations cannot test this because they vary all three factors at once. We isolate each factor in a controlled experiment across six benchmarks, four models from three families, and five enrichment levels, totaling over 24,000 evaluated responses. The assumption does not hold. Most enrichment reduces accuracy. Models prompted to use confidence scores comply correctly yet produce worse answers, a gap between utilization and accuracy that no prior work has measured. What determines answer quality is not how much metadata the context carries but whether the model can act on it for the given task. When metadata and retrieval strategy are aligned with model capabilities, a smaller model outperforms a frontier model by 19 F1 points. These findings motivate a processability hierarchy that predicts, from pre-training properties alone, which metadata a model can productively use, reframing RAG design as a question of model-context alignment rather than metadata accumulation.