mlsanityData to 7 Oct 2026

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

arXiv 2406.12824 ↗First posted 18 Jun 2024cs.CL cs.AIcited by 16

On the map: Retrieval-augmented generation → RAG robustness and adaptive retrieval

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.

Key facts. Cited by 16 papers in the map (data through 7 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.

Builds on

18 of 31 references.

Closest ancestors · 1

Same direction, newest first

  1. 2022Improving the Domain Adaptation of Retrieval Augmented Generation (RAG) Models for Open Domain Question Answering6 Oct

Other references · 17

  1. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  2. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  3. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  4. 2022Locating and Editing Factual Associations in GPTcited by 2,609
  5. 2022Mass-Editing Memory in a Transformercited by 976
  6. 2023Towards Automated Circuit Discovery for Mechanistic Interpretabilitycited by 700
  7. 2021Retrieval Augmentation Reduces Hallucination in Conversationcited by 691
  8. 2023Textbooks Are All You Need II: phi-1.5 technical reportcited by 559
  9. 2023Dissecting Recall of Factual Associations in Auto-Regressive Language Modelscited by 512
  10. 2024Retrieval Head Mechanistically Explains Long-Context Factualitycited by 219
  11. 2023Investigating the Factual Knowledge Boundary of Large Language Models with Retrieval Augmentationcited by 146
  12. 2024Model Editing at Scale leads to Gradual and Catastrophic Forgettingcited by 92
5 more
  1. 2024Detoxifying Large Language Models via Knowledge Editingcited by 86
  2. 2024A Unified Framework for Model Editingcited by 54
  3. 2024Locating and Editing Factual Associations in Mambacited by 35
  4. 2024ERAGent: Enhancing Retrieval-Augmented Language Models with Improved Accuracy, Efficiency, and Personalizationcited by 15
  5. 2023The Effect of Scaling, Retrieval Augmentation and Form on the Factual Consistency of Language Modelscited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 5; 2026: 6.

16 papers.

Sort:

By area:

  1. 2024ReDeEP: Detecting Hallucination in Retrieval-Augmented Generation via Mechanistic Interpretabilitycited by 100
  2. 2024Is Semantic Chunking Worth the Computational Cost?cited by 43
  3. 2025A Survey on Knowledge-Oriented Retrieval-Augmented Generationcited by 34
  4. 2025Shifting from Ranking to Set Selection for Retrieval Augmented Generationcited by 13
  5. 2024Investigating Context-Faithfulness in Large Language Models: The Roles of Memory Strength and Evidence Stylecited by 11
  6. 2025LLM Microscope: What Model Internals Reveal About Answer Correctness and Context Utilizationcited by 11
  7. 2024Deciphering the Interplay of Parametric and Non-parametric Memory in Retrieval-augmented Language Modelscited by 10
  8. 2025SUGAR: Leveraging Contextual Confidence for Smarter Retrievalcited by 6
  9. 2026NOVA: NOise-aware Verbal Confidence CAlibration for Robust Large Language Models in RAG Systemscited by 5
  10. 2024Familiarity-Aware Evidence Compression for Retrieval-Augmented Generationcited by 4
  11. 2026Taxonomy of the Retrieval System Framework: Pitfalls and Paradigmscited by 4
  12. 2025Can LLMs Replace Humans During Code Chunking?cited by 2
4 more
  1. 2026How Retrieved Context Shapes Internal Representations in RAGcited by 2
  2. 2026Parametric Memory Decoding for Zero-Shot Routing in LoRA-Based External Parametric Memorycited by 2
  3. 2026To Copy or Not to Copy: Copying Is Easier to Induce Than Recallcited by 1
  4. 2026The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Quantifying reliance on external information over parametric knowledge during Retrieval Augmented Generation (RAG) using mechanistic analysis
  2. 2025Retrieval-Augmented Generation: A Comprehensive Survey of Architectures, Enhancements, and Robustness Frontiers
  3. 2025Understanding Parametric Knowledge Injection in Retrieval-Augmented Generation
  4. 2024Context Awareness Gate For Retrieval Augmented Generation
  5. 2024The Power of Noise: Redefining Retrieval for RAG Systems
  6. 2024Do You Know What You Are Talking About? Characterizing Query-Knowledge Relevance For Reliable Retrieval Augmented Generation
  7. 2026To Memorize or to Retrieve: Scaling the Interaction Between Pretraining and Retrieval
  8. 2025Enhancing Retrieval-Augmented Generation: A Study of Best Practices