mlsanityData to 7 Oct 2026

Tug-of-War Between Knowledge: Exploring and Resolving Knowledge Conflicts in Retrieval-Augmented Language Models

arXiv 2402.14409 ↗First posted 22 Feb 2024cs.CL cs.AI cs.IRcited by 93

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

Abstract (arXiv)

Retrieval-augmented language models (RALMs) have demonstrated significant potential in refining and expanding their internal memory by retrieving evidence from external sources. However, RALMs will inevitably encounter knowledge conflicts when integrating their internal memory with external sources. Knowledge conflicts can ensnare RALMs in a tug-of-war between knowledge, limiting their practical applicability. In this paper, we focus on exploring and resolving knowledge conflicts in RALMs. First, we present an evaluation framework for assessing knowledge conflicts across various dimensions. Then, we investigate the behavior and preference of RALMs from the following two perspectives: (1) Conflicts between internal memory and external sources: We find that stronger RALMs emerge with the Dunning-Kruger effect, persistently favoring their faulty internal memory even when correct evidence is provided. Besides, RALMs exhibit an availability bias towards common knowledge; (2) Conflicts between truthful, irrelevant and misleading evidence: We reveal that RALMs follow the principle of majority rule, leaning towards placing trust in evidence that appears more frequently. Moreover, we find that RALMs exhibit confirmation bias, and are more willing to choose evidence that is consistent with their internal memory. To solve the challenge of knowledge conflicts, we propose a method called Conflict-Disentangle Contrastive Decoding (CD2) to better calibrate the model's confidence. Experimental results demonstrate that our CD2 can effectively resolve knowledge conflicts in RALMs.

Key facts. Cited by 93 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 29 of its 38 references that are in the map. First posted on arXiv 22 Feb 2024.

Builds on

29 of 38 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Benchmarking Large Language Models in Retrieval-Augmented Generation4 Sep

Other references · 28

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  3. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  4. 2022Scaling Instruction-Finetuned Language Modelscited by 2,997
  5. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  6. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,126
  7. 2021Improving language models by retrieving from trillions of tokenscited by 1,278
  8. 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,103
  9. 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 893
  10. 2023REPLUG: Retrieval-Augmented Black-Box Language Modelscited by 786
  11. 2023Large Language Modelscited by 774
  12. 2023In-Context Retrieval-Augmented Language Modelscited by 694
16 more
  1. 2023Baichuan 2: Open Large-scale Language Modelscited by 664
  2. 2022Contrastive Decoding: Open-ended Text Generation as Optimizationcited by 598
  3. 2020KILT: a Benchmark for Knowledge Intensive Language Taskscited by 571
  4. 2021Entity-Based Knowledge Conflicts in Question Answeringcited by 349
  5. 2023Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflictscited by 333
  6. 2023Trusting Your Evidence: Hallucinate Less with Context-aware Decodingcited by 319
  7. 2022Generate rather than Retrieve: Large Language Models are Strong Context Generatorscited by 296
  8. 2023Physics of Language Models: Part 3.1, Knowledge Storage and Extractioncited by 264
  9. 2023Head-to-Tail: How Knowledgeable are Large Language Models (LLMs)? A.K.A. Will LLMs Replace Knowledge Graphs?cited by 167
  10. 2023Evaluating Correctness and Faithfulness of Instruction-Following Models for Question Answeringcited by 150
  11. 2023Investigating the Factual Knowledge Boundary of Large Language Models with Retrieval Augmentationcited by 146
  12. 2022Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidencecited by 142
  13. 2022TemporalWiki: A Lifelong Benchmark for Training and Evaluating Ever-Evolving Language Modelscited by 117
  14. 2022DisentQA: Disentangling Parametric and Contextual Knowledge with Counterfactual Question Answeringcited by 93
  15. 2023"Merge Conflicts!" Exploring the Impacts of External Distractors to Parametric Knowledge Graphscited by 11
  16. 2023The Confidence-Competence Gap in Large Language Models: A Cognitive Studycited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 23; 2025: 44; 2026: 26.

93 papers.

Sort:

By area:

  1. 2024Knowledge Conflicts for LLMs: A Surveycited by 274
  2. 2024Retrieval-Augmented Generation with Graphs (GraphRAG)cited by 170
  3. 2024Cutting Off the Head Ends the Conflict: A Mechanism for Interpreting and Mitigating Knowledge Conflicts in Language Modelscited by 87
  4. 2025SealQA: Raising the Bar for Reasoning in Search-Augmented Language Modelscited by 65
  5. 2025Retrieval-Augmented Generation with Conflicting Evidencecited by 60
  6. 2024Astute RAG: Overcoming Imperfect Retrieval Augmentation and Knowledge Conflicts for Large Language Modelscited by 57
  7. 2024Knowledge Mechanisms in Large Language Models: A Survey and Perspectivecited by 50
  8. 2025LLM-based Agents Suffer from Hallucinations: A Survey of Taxonomy, Methods, and Directionscited by 49
  9. 2024WikiContradict: A Benchmark for Evaluating LLMs on Real-World Knowledge Conflicts from Wikipediacited by 43
  10. 2024RAG-DDR: Optimizing Retrieval-Augmented Generation Using Differentiable Data Rewardscited by 33
  11. 2025A Systematic Review of Key Retrieval-Augmented Generation (RAG) Systems: Progress, Gaps, and Future Directionscited by 30
  12. 2025Taming Knowledge Conflicts in Language Modelscited by 27
38 more
  1. 2024CrAM: Credibility-Aware Attention Modification in LLMs for Combating Misinformation in RAGcited by 22
  2. 2025Enhancing Retrieval-Augmented Generation: A Study of Best Practicescited by 22
  3. 2024KnowPO: Knowledge-aware Preference Optimization for Controllable Knowledge Selection in Retrieval-Augmented Language Modelscited by 20
  4. 2025Chain-of-Thought Prompting Obscures Hallucination Cues in Large Language Models: An Empirical Evaluationcited by 20
  5. 2024MRAG: A Modular Retrieval Framework for Time-Sensitive Question Answeringcited by 17
  6. 2025Your AI, Not Your View: The Bias of LLMs in Investment Analysiscited by 16
  7. 2025Task Matters: Knowledge Requirements Shape LLM Responses to Context-Memory Conflictcited by 15
  8. 2024Parenting: Optimizing Knowledge Selection of Retrieval-Augmented Language Models with Parameter Decoupling and Tailored Tuningcited by 14
  9. 2025Epistemic Alignment: A Mediating Framework for User-LLM Knowledge Deliverycited by 14
  10. 2024LLMs' Reading Comprehension Is Affected by Parametric Knowledge and Struggles with Hypothetical Statementscited by 12
  11. 2024Failure Modes of LLMs for Causal Reasoning on Narrativescited by 12
  12. 2025Firm or Fickle? Evaluating Large Language Models Consistency in Sequential Interactionscited by 12
  13. 2025TruthfulRAG: Resolving Factual-level Conflicts in Retrieval-Augmented Generation with Knowledge Graphscited by 12
  14. 2024Investigating Context-Faithfulness in Large Language Models: The Roles of Memory Strength and Evidence Stylecited by 11
  15. 2025ParamMute: Suppressing Knowledge-Critical FFNs for Faithful Retrieval-Augmented Generationcited by 11
  16. 2024Context-Parametric Inversion: Why Instruction Finetuning Can Worsen Context Reliancecited by 9
  17. 2025Dynamic Attention-Guided Context Decoding for Mitigating Context Faithfulness Hallucinations in Large Language Modelscited by 9
  18. 2025Systematic Knowledge Injection into Large Language Models via Diverse Augmentation for Domain-Specific RAGcited by 9
  19. 2026Whose Facts Win? LLM Source Preferences under Knowledge Conflictscited by 8
  20. 2024A Theory for Token-Level Harmonization in Retrieval-Augmented Generationcited by 7
  21. 2024Faux Polyglot: A Study on Information Disparity in Multilingual Large Language Modelscited by 7
  22. 2024ChroKnowledge: Unveiling Chronological Knowledge of Language Models in Multiple Domainscited by 7
  23. 2025Generate, Discriminate, Evolve: Enhancing Context Faithfulness via Fine-Grained Sentence-Level Self-Evolutioncited by 7
  24. 2025Can LLMs Reconcile Knowledge Conflicts in Counterfactual Reasoningcited by 7
  25. 2024Not All Languages are Equal: Insights into Multilingual Retrieval-Augmented Generationcited by 6
  26. 2025Rethinking All Evidence: Enhancing Trustworthy Retrieval-Augmented Generation via Conflict-Driven Summarizationcited by 6
  27. 2026Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generationcited by 6
  28. 2024Towards a Robust Retrieval-Based Summarization Systemcited by 5
  29. 2025How does Misinformation Affect Large Language Model Behaviors and Preferences?cited by 5
  30. 2025Parametric Retrieval-Augmented Generation using Latent Routing of LoRA Adapterscited by 5
  31. 2026Investigating Tool-Memory Conflicts in Tool-Augmented LLMscited by 5
  32. 2025MODS: Moderating a Mixture of Document Speakers to Summarize Debatable Queries in Document Collectionscited by 4
  33. 2025Accommodate Knowledge Conflicts in Retrieval-augmented LLMs: Towards Robust Response Generation in the Wildcited by 4
  34. 2025Transparentize the Internal and External Knowledge Utilization in LLMs with Trustworthy Citationcited by 4
  35. 2025Unveiling Knowledge Utilization Mechanisms in LLM-based Retrieval-Augmented Generationcited by 4
  36. 2025Exploring the Impact of Instruction-Tuning on LLM's Susceptibility to Misinformationcited by 4
  37. 2025Continuously Steering LLMs Sensitivity to Contextual Knowledge with Proxy Modelscited by 4
  38. 2025Detecting Corpus-Level Knowledge Inconsistencies in Wikipedia with Large Language Modelscited by 4

50 most cited of 93 All 93 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Exploring Knowledge Conflicts for Faithful LLM Reasoning: Benchmark and Method
  2. 2024DYNAMICQA: Tracing Internal Knowledge Conflicts in Language Models
  3. 2025FaithfulRAG: Fact-Level Conflict Modeling for Context-Faithful Retrieval-Augmented Generation
  4. 2026Trust or Abstain? A Self-Aware RAG Approach
  5. 2024Studying Large Language Model Behaviors Under Context-Memory Conflicts With Real Documents
  6. 2025DRAGged into Conflicts: Detecting and Addressing Conflicting Sources in Search-Augmented LLMs
  7. 2026Large Language Models in Resolving Contextual Knowledge Conflicts
  8. 2024ClashEval: Quantifying the tug-of-war between an LLM's internal prior and external evidence