mlsanityData to 7 Oct 2026

DisentQA: Disentangling Parametric and Contextual Knowledge with Counterfactual Question Answering

arXiv 2211.05655 ↗First posted 10 Nov 2022cs.CL cs.AI cs.LGcited by 93

Not on the map. Why?

Abstract (arXiv)

Question answering models commonly have access to two sources of "knowledge" during inference time: (1) parametric knowledge - the factual knowledge encoded in the model weights, and (2) contextual knowledge - external knowledge (e.g., a Wikipedia passage) given to the model to generate a grounded answer. Having these two sources of knowledge entangled together is a core issue for generative QA models as it is unclear whether the answer stems from the given non-parametric knowledge or not. This unclarity has implications on issues of trust, interpretability and factuality. In this work, we propose a new paradigm in which QA models are trained to disentangle the two sources of knowledge. Using counterfactual data augmentation, we introduce a model that predicts two answers for a given question: one based on given contextual knowledge and one based on parametric knowledge. Our experiments on the Natural Questions dataset show that this approach improves the performance of QA models by making them more robust to knowledge conflicts between the two knowledge sources, while generating useful disentangled answers.

Key facts. Cited by 93 papers in the map (data through 7 Oct 2026). Builds on 23 of its 32 references that are in the map. First posted on arXiv 10 Nov 2022.

Builds on

23 of 32 references.

References · 23

  1. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  2. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,115
  3. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,919
  4. 2022Locating and Editing Factual Associations in GPTcited by 2,609
  5. 2019BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questionscited by 2,140
  6. 2019Language Models as Knowledge Bases?cited by 1,922
  7. 2018Know What You Don't Know: Unanswerable Questions for SQuADcited by 1,538
  8. 2020How Much Knowledge Can You Pack Into the Parameters of a Language Model?cited by 713
  9. 2021Knowledge Neurons in Pretrained Transformerscited by 590
  10. 2020KILT: a Benchmark for Knowledge Intensive Language Taskscited by 571
  11. 2021Editing Factual Knowledge in Language Modelscited by 548
  12. 2021Entity-Based Knowledge Conflicts in Question Answeringcited by 349
11 more
  1. 2021Time-Aware Language Models as Temporal Knowledge Basescited by 290
  2. 2021Measuring Attribution in Natural Language Generation Modelscited by 253
  3. 2020Modifying Memories in Transformer Modelscited by 222
  4. 2022Scaling Up Models and Data with textttt5x and textttseqiocited by 187
  5. 2022Large Language Models with Controllable Working Memorycited by 169
  6. 2022Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidencecited by 142
  7. 2020Question and Answer Test-Train Overlap in Open-Domain Question Answering Datasetscited by 137
  8. 2022Investigating Selective Prediction Approaches Across Several Tasks in IID, OOD, and Adversarial Settingscited by 66
  9. 2020Challenges in Information-Seeking QA: Unanswerable Questions and Paragraph Retrievalcited by 44
  10. 2021Which Linguist Invented the Lightbulb? Presupposition Verification for Question-Answeringcited by 41
  11. 2018A Qualitative Comparison of CoQA, SQuAD 2.0 and QuACcited by 21

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 20; 2024: 40; 2025: 16; 2026: 17.

93 papers.

Sort:

By area:

  1. 2023Making Retrieval-Augmented Language Models Robust to Irrelevant Contextcited by 364
  2. 2024Knowledge Conflicts for LLMs: A Surveycited by 274
  3. 2023Survey on Factuality in Large Language Models: Knowledge, Retrieval and Domain-Specificitycited by 204
  4. 2024Lookback Lens: Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Mapscited by 140
  5. 2024To Believe or Not to Believe Your LLMcited by 120
  6. 2024Know Your Limits: A Survey of Abstention in Large Language Modelscited by 116
  7. 2023LawBench: Benchmarking Legal Knowledge of Large Language Modelscited by 111
  8. 2024Tug-of-War Between Knowledge: Exploring and Resolving Knowledge Conflicts in Retrieval-Augmented Language Modelscited by 93
  9. 2024Cutting Off the Head Ends the Conflict: A Mechanism for Interpreting and Mitigating Knowledge Conflicts in Language Modelscited by 87
  10. 2023LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugincited by 82
  11. 2023Context-faithful Prompting for Large Language Modelscited by 76
  12. 2023Factually Consistent Summarization via Reinforcement Learning with Textual Entailment Feedbackcited by 71
38 more
  1. 2024Learning to Route LLMs with Confidence Tokenscited by 66
  2. 2023Why So Gullible? Enhancing the Robustness of Retrieval-Augmented Models against Counterfactual Noisecited by 52
  3. 2023Resolving Knowledge Conflicts in Large Language Modelscited by 49
  4. 2024Enhancing Contextual Understanding in Large Language Models through Contrastive Decodingcited by 49
  5. 2024VLind-Bench: Measuring Language Priors in Large Vision-Language Modelscited by 44
  6. 2025Measuring Chain of Thought Faithfulness by Unlearning Reasoning Stepscited by 43
  7. 2023Merging Generated and Retrieved Knowledge for Open-Domain QAcited by 36
  8. 2024DYNAMICQA: Tracing Internal Knowledge Conflicts in Language Modelscited by 34
  9. 2023How Do Large Language Models Capture the Ever-changing World Knowledge? A Review of Recent Advancescited by 33
  10. 2023HallE-Control: Controlling Object Hallucination in Large Multimodal Modelscited by 31
  11. 2024Genie: Achieving Human Parity in Content-Grounded Datasets Generationcited by 31
  12. 2023Trends in Integration of Knowledge and Large Language Models: A Survey and Taxonomy of Methods, Benchmarks, and Applicationscited by 28
  13. 2025ECLeKTic: a Novel Challenge Set for Evaluation of Cross-Lingual Knowledge Transfercited by 28
  14. 2023Emptying the Ocean with a Spoon: Should We Edit Models?cited by 27
  15. 2024Discerning and Resolving Knowledge Conflicts through Adaptive Decoding with Contextual Information-Entropy Constraintcited by 27
  16. 2024Cofca: A Step-Wise Counterfactual Multi-hop QA benchmarkcited by 26
  17. 2023IfQA: A Dataset for Open-domain Question Answering under Counterfactual Presuppositionscited by 25
  18. 2023A Glitch in the Matrix? Locating and Detecting Language Model Grounding with Fakepediacited by 24
  19. 2024Claim Verification in the Age of Large Language Models: A Surveycited by 23
  20. 2024Controllable Context Sensitivity and the Knob Behind Itcited by 22
  21. 2024Who's Who: Large Language Models Meet Knowledge Conflicts in Practicecited by 20
  22. 2024Constructing Benchmarks and Interventions for Combating Hallucinations in LLMscited by 19
  23. 2024HFT: Half Fine-Tuning for Large Language Modelscited by 18
  24. 2023Assessing Knowledge Editing in Language Models via Relation Perspectivecited by 15
  25. 2024Unveiling Factual Recall Behaviors of Large Language Models through Knowledge Neuronscited by 15
  26. 2024Adaptive Question Answering: Enhancing Language Model Proficiency for Addressing Knowledge Conflicts with Source Citationscited by 15
  27. 2024DebateQA: Evaluating Question Answering on Debatable Knowledgecited by 12
  28. 2024Retrieving, Rethinking and Revising: The Chain-of-Verification Can Improve Retrieval Augmented Generationcited by 12
  29. 2023"Merge Conflicts!" Exploring the Impacts of External Distractors to Parametric Knowledge Graphscited by 11
  30. 2024Investigating Context-Faithfulness in Large Language Models: The Roles of Memory Strength and Evidence Stylecited by 11
  31. 2024CROPE: Evaluating In-Context Adaptation of Vision and Language Models to Culture-Specific Conceptscited by 11
  32. 2025ParamMute: Suppressing Knowledge-Critical FFNs for Faithful Retrieval-Augmented Generationcited by 11
  33. 2024When Context Leads but Parametric Memory Follows in Large Language Modelscited by 10
  34. 2024Utilize the Flow before Stepping into the Same River Twice: Certainty Represented Knowledge Flow for Refusal-Aware Instruction Tuningcited by 10
  35. 2025ConInstruct: Evaluating Large Language Models on Conflict Detection and Resolution in Instructionscited by 9
  36. 2023Robust Training for Conversational Question Answering Models with Reinforced Reformulation Generationcited by 7
  37. 2024Desiderata for the Context Use of Question Answering Systemscited by 7
  38. 2024The Queen of England is not England's Queen: On the Lack of Factual Coherency in PLMscited by 7

50 most cited of 93 All 93 (JSON)