mlsanityData to 7 Oct 2026

R-Tuning: Instructing Large Language Models to Say `I Don't Know'

arXiv 2311.09677 ↗First posted 16 Nov 2023cs.CLcited by 106

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) have revolutionized numerous domains with their impressive performance but still face their challenges. A predominant issue is the propensity for these models to generate non-existent facts, a concern termed hallucination. Our research is motivated by the observation that previous instruction tuning methods force the model to complete a sentence no matter whether the model knows the knowledge or not. When the question is out of the parametric knowledge, it will try to make up something and fail to indicate when it lacks knowledge. In this paper, we present a new approach called Refusal-Aware Instruction Tuning (R-Tuning). This approach is formalized by first identifying the disparity in knowledge encompassed by pre-trained parameters compared to that of instruction tuning data. Then, we construct the refusal-aware data based on the knowledge intersection, to tune LLMs to refrain from responding to questions beyond its parametric knowledge. Experimental results demonstrate R-Tuning effectively improves a model's ability to answer known questions and refrain from answering unknown questions. Furthermore, when tested on out-of-domain datasets, the refusal ability was found to be a meta-skill that could be generalized to other tasks. Further analysis surprisingly finds that learning the uncertainty results in better calibration and an improved ability to estimate the uncertainty than uncertainty-based testing. Our code is available at https://github.com/shizhediao/R-Tuning.

Key facts. Cited by 106 papers in the map (data through 7 Oct 2026). Builds on 33 of its 52 references that are in the map. First posted on arXiv 16 Nov 2023.

Builds on

33 of 52 references.

References · 33

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  3. 2017On Calibration of Modern Neural Networkscited by 4,860
  4. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  5. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,919
  6. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,800
  7. 2022Language Models (Mostly) Know What They Knowcited by 1,783
  8. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
  9. 2022Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?cited by 1,413
  10. 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,362
  11. 2023LIMA: Less Is More for Alignmentcited by 1,080
  12. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 965
21 more
  1. 2023The RefinedWeb Dataset for Falcon LLM: Outperforming Curated Corpora with Web Data, and Web Data Onlycited by 755
  2. 2023CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquingcited by 751
  3. 2021Measuring and Improving Consistency in Pretrained Language Modelscited by 361
  4. 2023Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedbackcited by 351
  5. 2022Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matterscited by 221
  6. 2023A Stitch in Time Saves Nine: Detecting and Mitigating Hallucinations of LLMs by Validating Low-Confidence Generationcited by 202
  7. 2022Factuality Enhanced Language Models for Open-Ended Text Generationcited by 200
  8. 2022Large Language Models with Controllable Working Memorycited by 169
  9. 2023LM vs LM: Detecting Factual Errors via Cross Examinationcited by 155
  10. 2023Investigating the Factual Knowledge Boundary of Large Language Models with Retrieval Augmentationcited by 146
  11. 2022On the Origin of Hallucinations in Conversational Models: Is it the Datasets or the Models?cited by 137
  12. 2022StreamingQA: A Benchmark for Adaptation to New Knowledge over Time in Question Answering Modelscited by 123
  13. 2023WiCE: Real-World Entailment for Claims in Wikipediacited by 118
  14. 2023Question Decomposition Improves the Faithfulness of Model-Generated Reasoningcited by 107
  15. 2021Time Waits for No One! Analysis and Challenges of Temporal Misalignmentcited by 91
  16. 2023LMFlow: An Extensible Toolkit for Finetuning and Inference of Large Foundation Modelscited by 58
  17. 2023PURR: Efficiently Editing Language Model Hallucinations by Denoising Language Model Corruptionscited by 53
  18. 2023Won't Get Fooled Again: Answering Questions with False Premisescited by 47
  19. 2023Augmented Large Language Models with Parametric Knowledge Guidingcited by 37
  20. 2023Improving Retrieval-Augmented Large Language Models via Data Importance Learningcited by 17
  21. 2023Sequence-Level Certainty Reduces Hallucination In Knowledge-Grounded Dialogue Generationcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 27; 2025: 34; 2026: 44.

106 papers.

Sort:

By area:

  1. 2023Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervisioncited by 445
  2. 2024A Comprehensive Survey of Hallucination Mitigation Techniques in Large Language Modelscited by 266
  3. 2024Know Your Limits: A Survey of Abstention in Large Language Modelscited by 116
  4. 2024Trust or Escalate: LLM Judges with Provable Guarantees for Human Agreementcited by 99
  5. 2024Functional Benchmarks for Robust Evaluation of Reasoning Performance, and the Reasoning Gapcited by 79
  6. 2024Sufficient Context: A New Lens on Retrieval Augmented Generation Systemscited by 69
  7. 2024From Pixels to Insights: A Survey on Automatic Chart Understanding in the Era of Large Foundation Modelscited by 52
  8. 2025Rewarding Doubt: A Reinforcement Learning Approach to Calibrated Confidence Expression of Large Language Modelscited by 48
  9. 2025Calibrating Verbal Uncertainty as a Linear Feature to Reduce Hallucinationscited by 48
  10. 2024Learning or Self-aligning? Rethinking Instruction Fine-tuningcited by 43
  11. 2024Knowledge Overshadowing Causes Amalgamated Hallucination in Large Language Modelscited by 24
  12. 2025Metacognition and Uncertainty Communication in Humans and Large Language Modelscited by 23
38 more
  1. 2024A Survey on the Honesty of Large Language Modelscited by 21
  2. 2024Grounding and Evaluation for Large Language Models: Practical Challenges and Lessons Learned (Survey)cited by 20
  3. 2025TruthRL: Incentivizing Truthful LLMs via Reinforcement Learningcited by 20
  4. 2024Narrowing the Knowledge Evaluation Gap: Open-Domain Question Answering with Multi-Granularity Answerscited by 18
  5. 2025LoVeC: Reinforcement Learning for Better Verbalized Confidence in Long-Form Generationscited by 17
  6. 2024EVEDIT: Event-based Knowledge Editing with Deductive Editing Boundariescited by 16
  7. 2024Automatic Curriculum Expert Iteration for Reliable LLM Reasoningcited by 16
  8. 2025Refine Knowledge of Large Language Models via Adaptive Contrastive Learningcited by 14
  9. 2025Don't lie to your friends: Learning what you know from collaborative self-playcited by 14
  10. 2024Certainly Uncertain: A Benchmark and Metric for Multimodal Epistemic and Aleatoric Awarenesscited by 13
  11. 2024Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Modelscited by 12
  12. 2025Query-Level Uncertainty in Large Language Modelscited by 12
  13. 2024Cannot or Should Not? Automatic Analysis of Refusal Composition in IFT/RLHF Datasets and Refusal Behavior of Black-Box LLMscited by 11
  14. 2024MACAROON: Training Vision-Language Models To Be Your Engaged Partnerscited by 10
  15. 2024Fine-Tuning Large Language Models to Appropriately Abstain with Semantic Entropycited by 9
  16. 2025Don't Throw Away Your Pretrained Modelcited by 9
  17. 2025Do Large Language Models Know What They Are Capable Of?cited by 9
  18. 2026Causal Evidence that Language Models use Confidence to Drive Behaviorcited by 9
  19. 2024Fake Artificial Intelligence Generated Contents (FAIGC): A Survey of Theories, Detection Methods, and Opportunitiescited by 7
  20. 2024Forewarned is Forearmed: Leveraging LLMs for Data Synthesis through Failure-Inducing Explorationcited by 7
  21. 2024Fictitious Synthetic Data Can Improve LLM Factuality via Prerequisite Learningcited by 7
  22. 2024Epistemic Integrity in Large Language Modelscited by 7
  23. 2024Search Engines Post-ChatGPT: How Generative Artificial Intelligence Could Make Search Less Reliablecited by 6
  24. 2025Fast or Better? Balancing Accuracy and Cost in Retrieval-Augmented Generation with Flexible User Controlcited by 6
  25. 2025Disentangled Safety Adapters Enable Efficient Guardrails and Flexible Inference-Time Alignmentcited by 6
  26. 2025Loki's Dance of Illusions: A Comprehensive Survey of Hallucination in Large Language Modelscited by 6
  27. 2025Answer, Refuse, or Guess? Investigating Risk-Aware Decision Making in Language Modelscited by 5
  28. 2025None of the Above, Less of the Right: Parallel Patterns between Humans and LLMs on Multi-Choice Questions Answeringcited by 5
  29. 2025Can LLMs Refuse Questions They Do Not Know? Measuring Knowledge-Aware Refusal in Factual Taskscited by 5
  30. 2025Train for Truth, Keep the Skills: Binary Retrieval-Augmented Reward Mitigates Hallucinationscited by 5
  31. 2025AdaSearch: Balancing Parametric Knowledge and Search in Large Language Models via Reinforcement Learningcited by 5
  32. 2026Model-Adaptive Tool Necessity Reveals the Knowing-Doing Gap in LLM Tool Usecited by 5
  33. 2025Learning from Reference Answers: Versatile Language Model Alignment without Binary Human Preference Datacited by 4
  34. 2025Structural Entropy Guided Agent for Detecting and Repairing Knowledge Deficiencies in LLMscited by 4
  35. 2026Beyond Math and Code: Lightweight Corpus-Grounded Process Rewards for Factual Question Answeringcited by 4
  36. 2026Reinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMscited by 4
  37. 2024Financial Knowledge Large Language Modelcited by 3
  38. 2024Tell Me What You Don't Know: Enhancing Refusal Capabilities of Role-Playing Agents via Representation Space Analysis and Editingcited by 3

50 most cited of 106 All 106 (JSON)