mlsanityData to 14 Oct 2026

QueryBandits for Hallucination Mitigation: Exploiting Semantic Features for No-Regret Rewriting

arXiv 2508.16697 ↗First posted 22 Aug 2025cs.CL cs.AI cs.LGcited by 1

On the map: LLM reasoning → Hallucination detection and mitigation

Abstract (arXiv)

Advanced reasoning capabilities in Large Language Models (LLMs) have caused higher hallucination prevalence; yet most mitigation work focuses on after-the-fact filtering rather than shaping the queries that trigger them. We introduce QueryBandits, a bandit framework that designs rewrite strategies to maximize a reward model, that encapsulates hallucination propensity based upon the sensitivities of 17 linguistic features of the input query-and therefore, proactively steer LLMs away from generating hallucinations. Across 13 diverse QA benchmarks and 1,050 lexically perturbed queries per dataset, our top contextual QueryBandit (Thompson Sampling) achieves an 87.5% win rate over a no-rewrite baseline and also outperforms zero-shot static prompting ("paraphrase" or "expand") by 42.6% and 60.3% respectively. Therefore, we empirically substantiate the effectiveness of QueryBandits in mitigating hallucination via the intervention that takes the form of a query rewrite. Interestingly, certain static prompting strategies, which constitute a considerable number of current query rewriting literature, have a higher cumulative regret than the no-rewrite baseline, signifying that static rewrites can worsen hallucination. Moreover, we discover that the converged per-arm regression feature weight vectors substantiate that there is no single rewrite strategy optimal for all queries. In this context, guided rewriting via exploiting semantic features with QueryBandits can induce significant shifts in output behavior through forward-pass mechanisms, bypassing the need for retraining or gradient-based adaptation.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Hallucination detection and mitigation, one of 17 directions in LLM reasoning. Builds on 73 of its 114 references that are in the map. First posted on arXiv 22 Aug 2025.

Builds on

73 of 114 references.

References · 73

  1. 2017Attention Is All You Needcited by 60,361
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2017Proximal Policy Optimization Algorithmscited by 15,215
  4. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  5. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  6. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  7. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  8. 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,886
  9. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,148
  10. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  11. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  12. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
38 more
  1. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,975
  2. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,955
  3. 2019PIQA: Reasoning about Physical Commonsense in Natural Languagecited by 2,883
  4. 2018Deep contextualized word representationscited by 2,290
  5. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,212
  6. 2019BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questionscited by 2,183
  7. 2018Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answeringcited by 2,164
  8. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,729
  9. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,601
  10. 2018Know What You Don't Know: Unanswerable Questions for SQuADcited by 1,554
  11. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,407
  12. 2017Crowdsourcing Multiple Choice Science Questionscited by 765
  13. 2021BARTScore: Evaluating Generated Text as Text Generationcited by 722
  14. 2019MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalismscited by 662
  15. 2024Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexitycited by 451
  16. 2023Towards Measuring the Representation of Subjective Global Opinions in Language Modelscited by 437
  17. 2023Is ChatGPT a Good NLG Evaluator? A Preliminary Studycited by 422
  18. 2022Generate rather than Retrieve: Large Language Models are Strong Context Generatorscited by 298
  19. 2024A Comprehensive Survey of Hallucination Mitigation Techniques in Large Language Modelscited by 269
  20. 2019Negated and Misprimed Probes for Pretrained Language Models: Birds Can Talk, But Cannot Flycited by 243
  21. 2013(More) Efficient Reinforcement Learning via Posterior Samplingcited by 233
  22. 2022Factuality Enhanced Language Models for Open-Ended Text Generationcited by 202
  23. 2023ReMax: A Simple, Effective, and Efficient Reinforcement Learning Method for Aligning Large Language Modelscited by 187
  24. 2023LLM Lies: Hallucinations are not Bugs, but Features as Adversarial Examplescited by 177
  25. 2023GPT-NER: Named Entity Recognition via Large Language Modelscited by 160
  26. 2023Evaluating Correctness and Faithfulness of Instruction-Following Models for Question Answeringcited by 151
  27. 2023Controlled Decoding from Language Modelscited by 135
  28. 2023We're Afraid Language Models Aren't Modeling Ambiguitycited by 126
  29. 2023Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselvescited by 123
  30. 2023FollowBench: A Multi-level Fine-grained Constraints Following Benchmark for Large Language Modelscited by 117
  31. 2023Prompting and Evaluating Large Language Models for Proactive Dialogues: Clarification, Target-guided, and Non-collaborationcited by 116
  32. 2023Towards Mitigating Hallucination in Large Language Models via Self-Reflectioncited by 81
  33. 2023Language models are not naysayers: An analysis of language models on negation benchmarkscited by 74
  34. 2023Selectively Answering Ambiguous Questionscited by 66
  35. 2024PUB: A Pragmatics Understanding Benchmark for Assessing LLMs' Pragmatics Capabilitiescited by 59
  36. 2024Aligning Language Models to Explicitly Handle Ambiguitycited by 56
  37. 2020Efficient and Robust Algorithms for Adversarial Linear Contextual Banditscited by 43
  38. 2019Rare Words: A Major Problem for Contextualized Embeddings And How to Fix it by Attentive Mimickingcited by 41

50 most cited of 73 All 73 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Diagnosing and Repairing Factual Errors in RAG under Budget Constraintscited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2026No One Size Fits All: QueryBandits for Hallucination Mitigation
  2. 2024Rowen: Adaptive Retrieval-Augmented Generation for Hallucination Mitigation in LLMs
  3. 2025Theoretical Foundations and Mitigation of Hallucination in Large Language Models
  4. 2026Mitigating Prompt-Induced Hallucinations in Large Language Models via Structured Reasoning
  5. 2025Mitigating Hallucination in Large Language Models (LLMs): An Application-Oriented Survey on RAG, Reasoning, and Agentic Systems
  6. 2024THaMES: An End-to-End Tool for Hallucination Mitigation and Evaluation in Large Language Models
  7. 2026BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigation
  8. 2025Triggering Hallucinations in LLMs: A Quantitative Study of Prompt-Induced Hallucination in Large Language Models