mlsanityData to 7 Oct 2026

Blinded by Generated Contexts: How Language Models Merge Generated and Retrieved Contexts When Knowledge Conflicts?

arXiv 2401.11911 ↗First posted 22 Jan 2024cs.CL cs.AIcited by 47

Not on the map. Why?

Abstract (arXiv)

While auxiliary information has become a key to enhancing Large Language Models (LLMs), relatively little is known about how LLMs merge these contexts, specifically contexts generated by LLMs and those retrieved from external sources. To investigate this, we formulate a systematic framework to identify whether LLMs' responses are attributed to either generated or retrieved contexts. To easily trace the origin of the response, we construct datasets with conflicting contexts, i.e., each question is paired with both generated and retrieved contexts, yet only one of them contains the correct answer. Our experiments reveal a significant bias in several LLMs (GPT-4/3.5 and Llama2) to favor generated contexts, even when they provide incorrect information. We further identify two key factors contributing to this bias: i) contexts generated by LLMs typically show greater similarity to the questions, increasing their likelihood of being selected; ii) the segmentation process used in retrieved contexts disrupts their completeness, thereby hindering their full utilization in LLMs. Our analysis enhances the understanding of how LLMs merge diverse contexts, offers valuable insights for advancing current LLM augmentation methods, and highlights the risk of generated misinformation for retrieval-augmented LLMs.

Key facts. Cited by 47 papers in the map (data through 7 Oct 2026). Builds on 34 of its 40 references that are in the map. First posted on arXiv 22 Jan 2024.

Builds on

34 of 40 references.

References · 34

  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. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,462
  4. 2022Large Language Models are Zero-Shot Reasonerscited by 4,323
  5. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  6. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  7. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  8. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,185
  9. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,126
  10. 2019Language Models as Knowledge Bases?cited by 1,922
  11. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,203
  12. 2021Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivitycited by 1,122
22 more
  1. 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,103
  2. 2022Atlas: Few-shot Learning with Retrieval Augmented Language Modelscited by 940
  3. 2023REPLUG: Retrieval-Augmented Black-Box Language Modelscited by 786
  4. 2020How Much Knowledge Can You Pack Into the Parameters of a Language Model?cited by 713
  5. 2023In-Context Retrieval-Augmented Language Modelscited by 694
  6. 2021Editing Factual Knowledge in Language Modelscited by 548
  7. 2022Large Language Models Struggle to Learn Long-Tail Knowledgecited by 452
  8. 2023Making Retrieval-Augmented Language Models Robust to Irrelevant Contextcited by 364
  9. 2023Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflictscited by 333
  10. 2022Generate rather than Retrieve: Large Language Models are Strong Context Generatorscited by 296
  11. 2021Generated Knowledge Prompting for Commonsense Reasoningcited by 213
  12. 2023Chain-of-Note: Enhancing Robustness in Retrieval-Augmented Language Modelscited by 180
  13. 2021Retrieving and Reading: A Comprehensive Survey on Open-domain Question Answeringcited by 178
  14. 2023Investigating the Factual Knowledge Boundary of Large Language Models with Retrieval Augmentationcited by 146
  15. 2022Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidencecited by 142
  16. 2023On the Risk of Misinformation Pollution with Large Language Modelscited by 124
  17. 2022TemporalWiki: A Lifelong Benchmark for Training and Evaluating Ever-Evolving Language Modelscited by 117
  18. 2019Multi-passage BERT: A Globally Normalized BERT Model for Open-domain Question Answeringcited by 96
  19. 2023Beyond Factuality: A Comprehensive Evaluation of Large Language Models as Knowledge Generatorscited by 42
  20. 2023Augmented Large Language Models with Parametric Knowledge Guidingcited by 37
  21. 2023Neural Retrievers are Biased Towards LLM-Generated Contentcited by 37
  22. 2023Merging Generated and Retrieved Knowledge for Open-Domain QAcited by 36

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 4; 2025: 25; 2026: 18.

47 papers.

Sort:

By area:

  1. 2025Seeing but Not Believing: Probing the Disconnect Between Visual Attention and Answer Correctness in VLMscited by 48
  2. 2024AdaCAD: Adaptively Decoding to Balance Conflicts between Contextual and Parametric Knowledgecited by 35
  3. 2025Show Me the Work: Fact-Checkers' Requirements for Explainable Automated Fact-Checkingcited by 29
  4. 2025Taming Knowledge Conflicts in Language Modelscited by 27
  5. 2025FaithfulRAG: Fact-Level Conflict Modeling for Context-Faithful Retrieval-Augmented Generationcited by 27
  6. 2025DRAGged into Conflicts: Detecting and Addressing Conflicting Sources in Search-Augmented LLMscited by 21
  7. 2025Your AI, Not Your View: The Bias of LLMs in Investment Analysiscited by 16
  8. 2025TruthfulRAG: Resolving Factual-level Conflicts in Retrieval-Augmented Generation with Knowledge Graphscited by 12
  9. 2025The Atlas of In-Context Learning: How Attention Heads Shape In-Context Retrieval Augmentationcited by 11
  10. 2025Pre-training Limited Memory Language Models with Internal and External Knowledgecited by 11
  11. 2024A Reality Check on Context Utilisation for Retrieval-Augmented Generationcited by 9
  12. 2025CoCoA: Confidence and Context-Aware Adaptive Decoding for Resolving Knowledge Conflicts in Large Language Modelscited by 9
35 more
  1. 2025Rescuing the Unpoisoned: Efficient Defense against Knowledge Corruption Attacks on RAG Systemscited by 9
  2. 2026Whose Facts Win? LLM Source Preferences under Knowledge Conflictscited by 8
  3. 2025NeuroGenPoisoning: Neuron-Guided Attacks on Retrieval-Augmented Generation of LLM via Genetic Optimization of External Knowledgecited by 7
  4. 2025How does Misinformation Affect Large Language Model Behaviors and Preferences?cited by 5
  5. 2026Investigating Tool-Memory Conflicts in Tool-Augmented LLMscited by 5
  6. 2025CoCoA: Collaborative Chain-of-Agents for Parametric-Retrieved Knowledge Synergycited by 4
  7. 2025Continuously Steering LLMs Sensitivity to Contextual Knowledge with Proxy Modelscited by 4
  8. 2026GenArena: How Can We Achieve Human-Aligned Evaluation for Visual Generation Tasks?cited by 4
  9. 2024What External Knowledge is Preferred by LLMs? Characterizing and Exploring Chain of Evidence in Imperfect Context for Multi-Hop QAcited by 3
  10. 2025KScope: A Framework for Characterizing the Knowledge Status of Language Modelscited by 3
  11. 2025Format as a Prior: Quantifying and Analyzing Bias in LLMs for Heterogeneous Datacited by 3
  12. 2026Judging Against the Reference: Uncovering Knowledge-Driven Failures in LLM-Judges on QA Evaluationcited by 3
  13. 2025Conflicts in Texts: Data, Implications and Challengescited by 2
  14. 2024Improving Generated and Retrieved Knowledge Combination Through Zero-shot Generationcited by 1
  15. 2025SynDec: A Synthesize-then-Decode Approach for Arbitrary Textual Style Transfer via Large Language Modelscited by 1
  16. 2025ZoFia: Zero-Shot Fake News Detection with Entity-Guided Retrieval and Multi-LLM Interactioncited by 1
  17. 2026When Abundance Conceals Weakness: Knowledge Conflict in Multilingual Modelscited by 1
  18. 2026CompactRAG: Reducing LLM Calls and Token Overhead in Multi-Hop Question Answeringcited by 1
  19. 2026On the impact of retrieved content representations in RAG Pipelinescited by 1
  20. 2026EvoTrustRAG: Evolution-Aware Conflict Attribution and Evidence Handling for Reliable Retrieval-Augmented Generationcited by 1
  21. 2025On the Diminishing Returns of Complex Robust RAG Training in the Era of Powerful LLMscited by 0
  22. 2025UniKnow: A Unified Framework for Reliable Language Model Behavior across Parametric and External Knowledgecited by 0
  23. 2025RDF-Based Structured Quality Assessment Representation of Multilingual LLM Evaluationscited by 0
  24. 2025OptAgent: Optimizing Query Rewriting for E-commerce via Multi-Agent Simulationcited by 0
  25. 2025From Retrieval to Generation: Unifying External and Parametric Knowledge for Medical Question Answeringcited by 0
  26. 2026Answering the Wrong Question: Reasoning Trace Inversion for Abstention in LLMscited by 0
  27. 2026Align Documents to Questions: Question-Oriented Document Rewriting for Retrieval-Augmented Generationcited by 0
  28. 2026FuzzAgent: Multi-Agent System for Evolutionary Library Fuzzingcited by 0
  29. 2026Explain Like I'm 5 or Whatever I Choose: Evaluating the Interactive Potential of Language Model Responsescited by 0
  30. 2026Retrieved But Not Reliable: A Survey on Attacks, and Defenses in Retrieval-Augmented Generationcited by 0
  31. 2026From Passive Response to Proactive Correction: Enhancing LLM Robustness Against Input Fact Perturbationscited by 0
  32. 2026CoVer: Conflict-Aware Claim Verificationcited by 0
  33. 2026KC-Bench: A Dynamic Interactive Benchmark for Evaluating Knowledge Conflicts in LLM Agentscited by 0
  34. 2026GraphEcho: Structural Redundancy and Evidence Provenance in LLM Graph Agentscited by 0
  35. 2026A Shared Taste for Model-Written Text: The Generator-by-Selector Matrices of "AI-AI Bias" Show No Detectable Own-Model Premiumcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025On the Consistency of Multilingual Context Utilization in Retrieval-Augmented Generation
  2. 2026Beyond the Parameters: A Technical Survey of Contextual Enrichment in Large Language Models: From In-Context Prompting to Causal Retrieval-Augmented Generation
  3. 2024Investigating Context-Faithfulness in Large Language Models: The Roles of Memory Strength and Evidence Style
  4. 2024ClashEval: Quantifying the tug-of-war between an LLM's internal prior and external evidence
  5. 2024Context-DPO: Aligning Language Models for Context-Faithfulness
  6. 2025"Lost-in-the-Later": Framework for Quantifying Contextual Grounding in Large Language Models
  7. 2025On the Influence of Context Size and Model Choice in Retrieval-Augmented Generation Systems
  8. 2024WikiContradict: A Benchmark for Evaluating LLMs on Real-World Knowledge Conflicts from Wikipedia