mlsanityData to 7 Oct 2026

Know Your Limits: A Survey of Abstention in Large Language Models

arXiv 2407.18418 ↗First posted 25 Jul 2024cs.CLcited by 116

Not on the map. Why?

Abstract (arXiv)

Abstention, the refusal of large language models (LLMs) to provide an answer, is increasingly recognized for its potential to mitigate hallucinations and enhance safety in LLM systems. In this survey, we introduce a framework to examine abstention from three perspectives: the query, the model, and human values. We organize the literature on abstention methods, benchmarks, and evaluation metrics using this framework, and discuss merits and limitations of prior work. We further identify and motivate areas for future research, such as whether abstention can be achieved as a meta-capability that transcends specific tasks or domains, and opportunities to optimize abstention abilities in specific contexts. In doing so, we aim to broaden the scope and impact of abstention methodologies in AI systems.

Key facts. Cited by 116 papers in the map (data through 7 Oct 2026). Builds on 157 of its 179 references that are in the map. First posted on arXiv 25 Jul 2024.

Builds on

157 of 179 references.

References · 157

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2022Training language models to follow instructions with human feedbackcited by 15,712
  3. 2017Proximal Policy Optimization Algorithmscited by 14,441
  4. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  5. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  6. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  7. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  8. 2020Scaling Laws for Neural Language Modelscited by 6,411
  9. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  10. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  11. 2017On Calibration of Modern Neural Networkscited by 4,860
  12. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
38 more
  1. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  2. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  3. 2023QLoRA: Efficient Finetuning of Quantized LLMscited by 3,355
  4. 2022Training Compute-Optimal Large Language Modelscited by 2,927
  5. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 2,914
  6. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,748
  7. 2022Survey of Hallucination in Natural Language Generationcited by 2,056
  8. 2022Language Models (Mostly) Know What They Knowcited by 1,783
  9. 2018Know What You Don't Know: Unanswerable Questions for SQuADcited by 1,538
  10. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,330
  11. 2023Jailbreaking Black Box Large Language Models in Twenty Queriescited by 1,326
  12. 2019PubMedQA: A Dataset for Biomedical Research Question Answeringcited by 1,271
  13. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,235
  14. 2023Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!cited by 1,106
  15. 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,103
  16. 2022Scaling Laws for Reward Model Overoptimizationcited by 1,056
  17. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 946
  18. 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 893
  19. 2023BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Datasetcited by 828
  20. 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 795
  21. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 788
  22. 2022Teaching Models to Express Their Uncertainty in Wordscited by 731
  23. 2018CoQA: A Conversational Question Answering Challengecited by 674
  24. 2023The Internal State of an LLM Knows When It's Lyingcited by 644
  25. 2023Safe RLHF: Safe Reinforcement Learning from Human Feedbackcited by 630
  26. 2023Baseline Defenses for Adversarial Attacks Against Aligned Language Modelscited by 608
  27. 2017Deep Learning Scaling is Predictable, Empiricallycited by 556
  28. 2023GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Promptscited by 520
  29. 2024JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Modelscited by 494
  30. 2020TyDi QA: A Benchmark for Information-Seeking Question Answering in Typologically Diverse Languagescited by 486
  31. 2023XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Modelscited by 477
  32. 2021A Dataset of Information-Seeking Questions and Answers Anchored in Research Paperscited by 475
  33. 2024WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMscited by 460
  34. 2023"Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Modelscited by 457
  35. 2023Benchmarking Large Language Models for News Summarizationcited by 444
  36. 2024The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructionscited by 441
  37. 2020How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answeringcited by 439
  38. 2023Catastrophic Jailbreak of Open-source LLMs via Exploiting Generationcited by 439

50 most cited of 157 All 157 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 6; 2025: 30; 2026: 80.

116 papers.

Sort:

By area:

  1. 2025AbstentionBench: Reasoning LLMs Fail on Unanswerable Questionscited by 129
  2. 2024Steering Language Model Refusal with Sparse Autoencoderscited by 89
  3. 2025TruthRL: Incentivizing Truthful LLMs via Reinforcement Learningcited by 20
  4. 2024Plurals: A System for Guiding LLMs Via Simulated Social Ensemblescited by 17
  5. 2025Don't lie to your friends: Learning what you know from collaborative self-playcited by 14
  6. 2024Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Modelscited by 12
  7. 2025Query-Level Uncertainty in Large Language Modelscited by 12
  8. 2025Geometric Uncertainty for Detecting and Correcting Hallucinations in LLMscited by 12
  9. 2025GRAIT: Gradient-Driven Refusal-Aware Instruction Tuning for Effective Hallucination Mitigationcited by 11
  10. 2024Utilize the Flow before Stepping into the Same River Twice: Certainty Represented Knowledge Flow for Refusal-Aware Instruction Tuningcited by 10
  11. 2025Too Good to be Bad: On the Failure of LLMs to Role-Play Villainscited by 10
  12. 2025TALES: A Taxonomy and Analysis of Cultural Representations in LLM-generated Storiescited by 10
38 more
  1. 2026Over-Searching in Search-Augmented Large Language Modelscited by 10
  2. 2024Fine-Tuning Large Language Models to Appropriately Abstain with Semantic Entropycited by 9
  3. 2025Do Large Language Models Know What They Are Capable Of?cited by 9
  4. 2026Causal Evidence that Language Models use Confidence to Drive Behaviorcited by 9
  5. 2026Agentic Abstention: Do Agents Know When to Stop Instead of Act?cited by 9
  6. 2025RefusalBench: Generative Evaluation of Selective Refusal in Grounded Language Modelscited by 8
  7. 2026AgentAbstain: Do LLM Agents Know When Not to Act?cited by 8
  8. 2025UPCORE: Utility-Preserving Coreset Selection for Balanced Unlearningcited by 7
  9. 2025Aligning Large Language Models for Faithful Integrity Against Opposing Argumentcited by 6
  10. 2026Knowing When to Quit: A Principled Framework for Dynamic Abstention in LLM Reasoningcited by 6
  11. 2024Beyond Introspection: Reinforcing Thinking via Externalist Behavioral Feedbackcited by 5
  12. 2025Can Video LLMs Refuse to Answer? Alignment for Answerability in Video Large Language Modelscited by 5
  13. 2025Abstain and Validate: A Dual-LLM Policy for Reducing Noise in Agentic Program Repaircited by 5
  14. 2025Train for Truth, Keep the Skills: Binary Retrieval-Augmented Reward Mitigates Hallucinationscited by 5
  15. 2026When Silence Is Golden: Can LLMs Learn to Abstain in Temporal QA and Beyond?cited by 5
  16. 2026Geometry-Calibrated Conformal Abstention for Language Modelscited by 5
  17. 2025Hallucinate Less by Thinking More: Aspect-Based Causal Abstention for Large Language Modelscited by 4
  18. 2026Feedback by Design: Understanding and Overcoming User Feedback Barriers in Conversational Agentscited by 4
  19. 2026Semantic Self-Distillation for Language Model Uncertaintycited by 4
  20. 2026Walking Through Uncertainty: An Empirical Study of Uncertainty Estimation for Audio-Aware Large Language Modelscited by 4
  21. 2026Confidence Calibration for Multimodal LLMs: An Empirical Study through Medical VQAcited by 4
  22. 2025Measuring Faithfulness and Abstention: An Automated Pipeline for Evaluating LLM-Generated 3-ply Case-Based Legal Argumentscited by 3
  23. 2025KScope: A Framework for Characterizing the Knowledge Status of Language Modelscited by 3
  24. 2025Honesty over Accuracy: Trustworthy Language Models through Reinforced Hesitationcited by 3
  25. 2025Explicit Abstention Knobs for Predictable Reliability in Video Question Answeringcited by 3
  26. 2026Steering to Say No: Configurable Refusal via Activation Steering in Vision Language Modelscited by 3
  27. 2026Semantic Layers for Reliable LLM-Powered Data Analytics: A Paired Benchmark of Accuracy and Hallucination Across Three Frontier Modelscited by 3
  28. 2026Position: agentic AI orchestration should be Bayes-consistentcited by 3
  29. 2026Evaluating Confidence-Gated Retrieval with Matched Trajectory Replaycited by 3
  30. 2025R2-KG: General-Purpose Dual-Agent Framework for Reliable Reasoning on Knowledge Graphscited by 2
  31. 2025MKA: Leveraging Cross-Lingual Consensus for Model Abstentioncited by 2
  32. 2025Longitudinal Monitoring of LLM Content Moderation of Social Issuescited by 2
  33. 2025Countermind: A Multi-Layered Security Architecture for Large Language Modelscited by 2
  34. 2026Hallucination is a Consequence of Space-Optimality: A Rate-Distortion Theorem for Membership Testingcited by 2
  35. 2026Fine-Tuning Language Models to Know What They Knowcited by 2
  36. 2026Know More, Know Clearer: A Meta-Cognitive Framework for Knowledge Augmentation in Large Language Modelscited by 2
  37. 2026VIBEPASS: Can Vibe Coders Really Pass the Vibe Check?cited by 2
  38. 2026Neurosymbolic Learning for Inference-Time Argumentationcited by 2

50 most cited of 116 All 116 (JSON)