# Parametric Retrieval-Augmented Generation using Latent Routing of LoRA Adapters

arXiv: https://arxiv.org/abs/2511.17044 · HTML: https://mlsanity.com/paper/2511.17044/

## Key facts

- Cited by 6 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 26 of its 44 references that are in the map.
- First posted on arXiv 21 Nov 2025.

## Placed in

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

## Builds on (most cited first, 26 of 26)

- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection](https://mlsanity.com/paper/2310.11511/) (2023)
- [Improving language models by retrieving from trillions of tokens](https://mlsanity.com/paper/2112.04426/) (2021)
- [Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Steps](https://mlsanity.com/paper/2011.01060/) (2020)
- [Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering](https://mlsanity.com/paper/2007.01282/) (2020)
- [When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories](https://mlsanity.com/paper/2212.10511/) (2022)
- [Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions](https://mlsanity.com/paper/2212.10509/) (2022)
- [Large Language Models Can Be Easily Distracted by Irrelevant Context](https://mlsanity.com/paper/2302.00093/) (2023)
- [Large Language Models Struggle to Learn Long-Tail Knowledge](https://mlsanity.com/paper/2211.08411/) (2022)
- [The Web as a Knowledge-base for Answering Complex Questions](https://mlsanity.com/paper/1803.06643/) (2018)
- [Making Retrieval-Augmented Language Models Robust to Irrelevant Context](https://mlsanity.com/paper/2310.01558/) (2023)
- [The Power of Noise: Redefining Retrieval for RAG Systems](https://mlsanity.com/paper/2401.14887/) (2024)
- [RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMs](https://mlsanity.com/paper/2407.02485/) (2024)
- [LongRAG: Enhancing Retrieval-Augmented Generation with Long-context LLMs](https://mlsanity.com/paper/2406.15319/) (2024)
- [Tug-of-War Between Knowledge: Exploring and Resolving Knowledge Conflicts in Retrieval-Augmented Language Models](https://mlsanity.com/paper/2402.14409/) (2024)
- [Blended RAG: Improving RAG (Retriever-Augmented Generation) Accuracy with Semantic Search and Hybrid Query-Based Retrievers](https://mlsanity.com/paper/2404.07220/) (2024)
- [Trustworthiness in Retrieval-Augmented Generation Systems: A Survey](https://mlsanity.com/paper/2409.10102/) (2024)
- [Typos that Broke the RAG's Back: Genetic Attack on RAG Pipeline by Simulating Documents in the Wild via Low-level Perturbations](https://mlsanity.com/paper/2404.13948/) (2024)
- [Improving Retrieval for RAG based Question Answering Models on Financial Documents](https://mlsanity.com/paper/2404.07221/) (2024)
- [Multi-Head Adapter Routing for Cross-Task Generalization](https://mlsanity.com/paper/2211.03831/) (2022)
- [LaRA: Benchmarking Retrieval-Augmented Generation and Long-Context LLMs -- No Silver Bullet for LC or RAG Routing](https://mlsanity.com/paper/2502.09977/) (2025)
- [Knowledge Editing through Chain-of-Thought](https://mlsanity.com/paper/2412.17727/) (2024)

## Built on by (most cited first, 6 of 6)

- [FCPRAG: Fusion-Controller Parametric Retrieval-Augmented Generation for Stable Multi-Passage LoRA Injection](https://mlsanity.com/paper/2608.21750/) (2026)
- [Agentic Conversational Search with Contextualized Reasoning via Reinforcement Learning](https://mlsanity.com/paper/2601.13115/) (2026)
- [CroSearch-R1: Better Leveraging Cross-lingual Knowledge for Retrieval-Augmented Generation](https://mlsanity.com/paper/2604.25182/) (2026)
- [Learning to Route Queries to Heads for Attention-based Re-ranking with Large Language Models](https://mlsanity.com/paper/2604.24608/) (2026)
- [A Storage-Retrieval Gap in Parametric Knowledge Graph Memory](https://mlsanity.com/paper/2608.25489/) (2026)
- [Towards In-Parameter Memory Augmentation for Large Language Models](https://mlsanity.com/paper/2610.08630/) (2026)

## Abstract (arXiv)

Parametric Retrieval-Augmented Generation (PRAG) is a RAG approach that integrates external knowledge directly into model parameters using a LoRA adapter, aiming at reducing the inference cost compared to traditional RAG. However, current PRAG approaches adopt a textitone-to-one document encoding scheme, using a dedicated LoRA adapter for each individual document. This scheme introduces two major limitations: 1) As the number of documents increases, there will be a prohibitive cost for training and storage. 2) The LoRA adapters may largely overlap due to the shared knowledge across documents, making the approach highly inefficient. To overcome these challenges, we propose the Poly-PRAG approach, which uses a small set of LoRA adapters that are able to encode more general knowledge. Each document can be encoded using a combination of them through a latent routing function. By jointly training the LoRA adapters and the latent routing function, each LoRA adapter is able to encode a shared part of the knowledge across documents, and the routing function can select the best combination of adapters for a document. Experimental results on four benchmarks demonstrate the effectiveness of the Poly-PRAG compared to other strong PRAG baselines. In addition, this approach reduces the storage requirement by avoiding the need to store a large number of LoRA adapters and offers a more efficient way to encode external knowledge into LLMs.