# From RAGs to rich parameters: Probing how language models utilize external knowledge over parametric information for factual queries

arXiv: https://arxiv.org/abs/2406.12824 · HTML: https://mlsanity.com/paper/2406.12824/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation.
- Builds on 18 of its 31 references that are in the map.
- First posted on arXiv 18 Jun 2024.

## Placed in

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

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

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Mass-Editing Memory in a Transformer](https://mlsanity.com/paper/2210.07229/) (2022)
- [Towards Automated Circuit Discovery for Mechanistic Interpretability](https://mlsanity.com/paper/2304.14997/) (2023)
- [Retrieval Augmentation Reduces Hallucination in Conversation](https://mlsanity.com/paper/2104.07567/) (2021)
- [Textbooks Are All You Need II: phi-1.5 technical report](https://mlsanity.com/paper/2309.05463/) (2023)
- [Dissecting Recall of Factual Associations in Auto-Regressive Language Models](https://mlsanity.com/paper/2304.14767/) (2023)
- [Retrieval Head Mechanistically Explains Long-Context Factuality](https://mlsanity.com/paper/2404.15574/) (2024)
- [Investigating the Factual Knowledge Boundary of Large Language Models with Retrieval Augmentation](https://mlsanity.com/paper/2307.11019/) (2023)
- [Improving the Domain Adaptation of Retrieval Augmented Generation (RAG) Models for Open Domain Question Answering](https://mlsanity.com/paper/2210.02627/) (2022)
- [Model Editing at Scale leads to Gradual and Catastrophic Forgetting](https://mlsanity.com/paper/2401.07453/) (2024)
- [Detoxifying Large Language Models via Knowledge Editing](https://mlsanity.com/paper/2403.14472/) (2024)
- [A Unified Framework for Model Editing](https://mlsanity.com/paper/2403.14236/) (2024)
- [Locating and Editing Factual Associations in Mamba](https://mlsanity.com/paper/2404.03646/) (2024)
- [ERAGent: Enhancing Retrieval-Augmented Language Models with Improved Accuracy, Efficiency, and Personalization](https://mlsanity.com/paper/2405.06683/) (2024)
- [The Effect of Scaling, Retrieval Augmentation and Form on the Factual Consistency of Language Models](https://mlsanity.com/paper/2311.01307/) (2023)

## Built on by (most cited first, 16 of 16)

- [ReDeEP: Detecting Hallucination in Retrieval-Augmented Generation via Mechanistic Interpretability](https://mlsanity.com/paper/2410.11414/) (2024)
- [Is Semantic Chunking Worth the Computational Cost?](https://mlsanity.com/paper/2410.13070/) (2024)
- [A Survey on Knowledge-Oriented Retrieval-Augmented Generation](https://mlsanity.com/paper/2503.10677/) (2025)
- [Shifting from Ranking to Set Selection for Retrieval Augmented Generation](https://mlsanity.com/paper/2507.06838/) (2025)
- [Investigating Context-Faithfulness in Large Language Models: The Roles of Memory Strength and Evidence Style](https://mlsanity.com/paper/2409.10955/) (2024)
- [LLM Microscope: What Model Internals Reveal About Answer Correctness and Context Utilization](https://mlsanity.com/paper/2510.04013/) (2025)
- [Deciphering the Interplay of Parametric and Non-parametric Memory in Retrieval-augmented Language Models](https://mlsanity.com/paper/2410.05162/) (2024)
- [SUGAR: Leveraging Contextual Confidence for Smarter Retrieval](https://mlsanity.com/paper/2501.04899/) (2025)
- [NOVA: NOise-aware Verbal Confidence CAlibration for Robust Large Language Models in RAG Systems](https://mlsanity.com/paper/2601.11004/) (2026)
- [Familiarity-Aware Evidence Compression for Retrieval-Augmented Generation](https://mlsanity.com/paper/2409.12468/) (2024)
- [Taxonomy of the Retrieval System Framework: Pitfalls and Paradigms](https://mlsanity.com/paper/2601.20131/) (2026)
- [Can LLMs Replace Humans During Code Chunking?](https://mlsanity.com/paper/2506.19897/) (2025)
- [How Retrieved Context Shapes Internal Representations in RAG](https://mlsanity.com/paper/2602.20091/) (2026)
- [Parametric Memory Decoding for Zero-Shot Routing in LoRA-Based External Parametric Memory](https://mlsanity.com/paper/2607.04118/) (2026)
- [To Copy or Not to Copy: Copying Is Easier to Induce Than Recall](https://mlsanity.com/paper/2601.12075/) (2026)
- [The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generation](https://mlsanity.com/paper/2605.05594/) (2026)

## Abstract (arXiv)

Retrieval Augmented Generation (RAG) enriches the ability of language models to reason using external context to augment responses for a given user prompt. This approach has risen in popularity due to practical applications in various applications of language models in search, question/answering, and chat-bots. However, the exact nature of how this approach works isn't clearly understood. In this paper, we mechanistically examine the RAG pipeline to highlight that language models take shortcut and have a strong bias towards utilizing only the context information to answer the question, while relying minimally on their parametric memory. We probe this mechanistic behavior in language models with: (i) Causal Mediation Analysis to show that the parametric memory is minimally utilized when answering a question and (ii) Attention Contributions and Knockouts to show that the last token residual stream do not get enriched from the subject token in the question, but gets enriched from other informative tokens in the context. We find this pronounced shortcut behaviour true across both LLaMa and Phi family of models.