mlsanityData to 14 Oct 2026

Won't Get Fooled Again: Answering Questions with False Premises

arXiv 2307.02394 ↗First posted 5 Jul 2023cs.CLcited by 48

Not on the map. Why?

Abstract (arXiv)

Pre-trained language models (PLMs) have shown unprecedented potential in various fields, especially as the backbones for question-answering (QA) systems. However, they tend to be easily deceived by tricky questions such as "How many eyes does the sun have?". Such frailties of PLMs often allude to the lack of knowledge within them. In this paper, we find that the PLMs already possess the knowledge required to rebut such questions, and the key is how to activate the knowledge. To systematize this observation, we investigate the PLMs' responses to one kind of tricky questions, i.e., the false premises questions (FPQs). We annotate a FalseQA dataset containing 2365 human-written FPQs, with the corresponding explanations for the false premises and the revised true premise questions. Using FalseQA, we discover that PLMs are capable of discriminating FPQs by fine-tuning on moderate numbers (e.g., 256) of examples. PLMs also generate reasonable explanations for the false premise, which serve as rebuttals. Further replaying a few general questions during training allows PLMs to excel on FPQs and general questions simultaneously. Our work suggests that once the rebuttal ability is stimulated, knowledge inside the PLMs can be effectively utilized to handle FPQs, which incentivizes the research on PLM-based QA systems.

Key facts. Cited by 48 papers in the map (data through 14 Oct 2026). Builds on 29 of its 41 references that are in the map. First posted on arXiv 5 Jul 2023.

Builds on

29 of 41 references.

References · 29

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  4. 2020Scaling Laws for Neural Language Modelscited by 6,586
  5. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,893
  6. 2018GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understandingcited by 4,899
  7. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,148
  8. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,512
  9. 2015VQA: Visual Question Answeringcited by 2,506
  10. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,141
  11. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,998
  12. 2022BLOOM: A 176B-Parameter Open-Access Multilingual Language Modelcited by 1,984
17 more
  1. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,729
  2. 2018Know What You Don't Know: Unanswerable Questions for SQuADcited by 1,554
  3. 2020Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inferencecited by 938
  4. 2017RACE: Large-scale ReAding Comprehension Dataset From Examinationscited by 861
  5. 2020Recipes for building an open-domain chatbotcited by 593
  6. 2020UnifiedQA: Crossing Format Boundaries With a Single QA Systemcited by 476
  7. 2021Pre-Trained Models: Past, Present and Futurecited by 319
  8. 2021Alignment of Language Agentscited by 168
  9. 2021Hurdles to Progress in Long-form Question Answeringcited by 167
  10. 2021Think you have Solved Direct-Answer Question Answering? Try ARC-DA, the Direct-Answer AI2 Reasoning Challengecited by 80
  11. 2021Towards Interpreting and Mitigating Shortcut Learning Behavior of NLU Modelscited by 78
  12. 2022CREPE: Open-Domain Question Answering with False Presuppositionscited by 72
  13. 2020GenericsKB: A Knowledge Base of Generic Statementscited by 51
  14. 2020Challenges in Information-Seeking QA: Unanswerable Questions and Paragraph Retrievalcited by 44
  15. 2021Which Linguist Invented the Lightbulb? Presupposition Verification for Question-Answeringcited by 41
  16. 2021General-Purpose Question-Answering with Macawcited by 40
  17. 2021Unanswerable Questions about Images and Textscited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 3; 2024: 11; 2025: 19; 2026: 15.

48 papers.

Sort:

By area:

  1. 2024Tulu 3: Pushing Frontiers in Open Language Model Post-Trainingcited by 1,053
  2. 2023UltraFeedback: Boosting Language Models with Scaled AI Feedbackcited by 406
  3. 2024Don't Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaborationcited by 180
  4. 2025AbstentionBench: Reasoning LLMs Fail on Unanswerable Questionscited by 145
  5. 2025Preference Leakage: A Contamination Problem in LLM-as-a-judgecited by 125
  6. 2023R-Tuning: Instructing Large Language Models to Say `I Don't Know'cited by 111
  7. 2024SaySelf: Teaching LLMs to Express Confidence with Self-Reflective Rationalescited by 105
  8. 2024InfoRM: Mitigating Reward Hacking in RLHF via Information-Theoretic Reward Modelingcited by 79
  9. 2024Self-Boosting Large Language Models with Synthetic Preference Datacited by 40
  10. 2023Examining LLMs' Uncertainty Expression Towards Questions Outside Parametric Knowledgecited by 32
  11. 2025Large Language Models Meet Symbolic Provers for Logical Reasoning Evaluationcited by 28
  12. 2024A Survey on the Honesty of Large Language Modelscited by 23
36 more
  1. 2025How to Protect Yourself from 5G Radiation? Investigating LLM Responses to Implicit Misinformationcited by 18
  2. 2025Don't Let It Hallucinate: Premise Verification via Retrieval-Augmented Logical Reasoningcited by 15
  3. 2026Agentic Abstention: Do Agents Know When to Stop Instead of Act?cited by 15
  4. 2026Abstain-R1: Calibrated Abstention and Post-Refusal Clarification via Verifiable RLcited by 14
  5. 2025MMBoundary: Advancing MLLM Knowledge Boundary Awareness through Reasoning Step Confidence Calibrationcited by 11
  6. 2026Over-Searching in Search-Augmented Large Language Modelscited by 11
  7. 2024KG-FPQ: Evaluating Factuality Hallucination in LLMs with Knowledge Graph-based False Premise Questionscited by 9
  8. 2025MultiHoax: A Dataset of Multi-hop False-Premise Questionscited by 9
  9. 2025RefusalBench: Generative Evaluation of Selective Refusal in Grounded Language Modelscited by 9
  10. 2024Toward Robust RALMs: Revealing the Impact of Imperfect Retrieval on Retrieval-Augmented Language Modelscited by 8
  11. 2025Well Begun is Half Done: Low-resource Preference Alignment by Weak-to-Strong Decodingcited by 7
  12. 2025Do What? Teaching Vision-Language-Action Models to Reject the Impossiblecited by 7
  13. 2024I Could've Asked That: Reformulating Unanswerable Questionscited by 6
  14. 2025SIMPLEMIX: Frustratingly Simple Mixing of Off- and On-policy Data in Language Model Preference Learningcited by 6
  15. 2024DogeRM: Equipping Reward Models with Domain Knowledge through Model Mergingcited by 5
  16. 2025Information-Theoretic Reward Modeling for Stable RLHF: Detecting and Mitigating Reward Hackingcited by 5
  17. 2025MAC-Tuning: LLM Multi-Compositional Problem Reasoning with Enhanced Knowledge Boundary Awarenesscited by 3
  18. 2025Reverse Engineering Human Preferences with Reinforcement Learningcited by 3
  19. 2025UAQFact: Evaluating Factual Knowledge Utilization of LLMs on Unanswerable Questionscited by 3
  20. 2026Knowing but Not Correcting: Routine Task Requests Suppress Factual Correction in LLMscited by 3
  21. 2024ELOQ: Resources for Enhancing LLM Detection of Out-of-Scope Questionscited by 2
  22. 2025Judge Before Answer: Can MLLM Discern the False Premise in Question?cited by 2
  23. 2026Certainty robustness: Evaluating LLM stability under self-challenging promptscited by 2
  24. 2025P-Aligner: Enabling Pre-Alignment of Language Models via Principled Instruction Synthesiscited by 1
  25. 2025Is On-Policy Data always the Best Choice for Direct Preference Optimization-based LM Alignment?cited by 1
  26. 2026Learning Ordinal Probabilistic Reward from Preferencescited by 1
  27. 2026Don't `Well, Actually' Me Unless You Know What You're Talking About: Weak Presupposition Verification Degrades General QA Performancecited by 1
  28. 2025Identifying and Answering Questions with False Assumptions: An Interpretable Approachcited by 0
  29. 2026MedRedFlag: Investigating how LLMs Redirect Misconceptions in Real-World Health Communicationcited by 0
  30. 2026Measuring Stability and Failure Behavior in Language Models Under Structured Perturbationscited by 0
  31. 2026From Passive Response to Proactive Correction: Enhancing LLM Robustness Against Input Fact Perturbationscited by 0
  32. 2026What Makes Agent Memory Useful for Reliable Unanswerable Question Handling?cited by 0
  33. 2026Recognition-Refusal Misalignment in LLMs: Why Models Answer Structurally Unanswerable Questionscited by 0
  34. 2026FPCO-Dialog: A Multi-Turn False-Premise Benchmark for Correction and Cooperation in Vision-Language Modelscited by 0
  35. 2026When Context Misleads: In-context Learning with Jurisdiction in Large Language Modelscited by 0
  36. 2026Verification Trap: Understanding Test-Time Selection Failures under False Premises in Code Generationcited by 0