mlsanityData to 7 Oct 2026

Benchmarking and Improving Generator-Validator Consistency of Language Models

arXiv 2310.01846 ↗First posted 3 Oct 2023cs.CL cs.LGcited by 43

Not on the map. Why?

Abstract (arXiv)

As of September 2023, ChatGPT correctly answers "what is 7+8" with 15, but when asked "7+8=15, True or False" it responds with "False". This inconsistency between generating and validating an answer is prevalent in language models (LMs) and erodes trust. In this paper, we propose a framework for measuring the consistency between generation and validation (which we call generator-validator consistency, or GV-consistency), finding that even GPT-4, a state-of-the-art LM, is GV-consistent only 76% of the time. To improve the consistency of LMs, we propose to finetune on the filtered generator and validator responses that are GV-consistent, and call this approach consistency fine-tuning. We find that this approach improves GV-consistency of Alpaca-30B from 60% to 93%, and the improvement extrapolates to unseen tasks and domains (e.g., GV-consistency for positive style transfers extrapolates to unseen styles like humor). In addition to improving consistency, consistency fine-tuning improves both generator quality and validator accuracy without using any labeled data. Evaluated across 6 tasks, including math questions, knowledge-intensive QA, and instruction following, our method improves the generator quality by 16% and the validator accuracy by 6.3% across all tasks.

Key facts. Cited by 43 papers in the map (data through 7 Oct 2026). Builds on 20 of its 29 references that are in the map. First posted on arXiv 3 Oct 2023.

Builds on

20 of 29 references.

References · 20

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  3. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  4. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,878
  5. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
  6. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  7. 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,579
  8. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
  9. 2022Red Teaming Language Models with Language Modelscited by 983
  10. 2019CTRL: A Conditional Transformer Language Model for Controllable Generationcited by 754
  11. 2022Discovering Latent Knowledge in Language Models Without Supervisioncited by 733
  12. 2022Teaching Models to Express Their Uncertainty in Wordscited by 731
8 more
  1. 2021Measuring and Improving Consistency in Pretrained Language Modelscited by 361
  2. 2018Delete, Retrieve, Generate: A Simple Approach to Sentiment and Style Transfercited by 191
  3. 2020Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognitioncited by 185
  4. 2022Maieutic Prompting: Logically Consistent Reasoning with Recursive Explanationscited by 159
  5. 2021A Recipe For Arbitrary Text Style Transfer with Large Language Modelscited by 107
  6. 2022Prompt Consistency for Zero-Shot Task Generalizationcited by 66
  7. 2023Evaluating Superhuman Models with Consistency Checkscited by 47
  8. 2022Enhancing Self-Consistency and Performance of Pre-Trained Language Models through Natural Language Inferencecited by 40

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 16; 2025: 14; 2026: 11.

43 papers.

Sort:

By area:

  1. 2024PrivacyLens: Evaluating Privacy Norm Awareness of Language Models in Actioncited by 161
  2. 2024Rethinking Interpretability in the Era of Large Language Modelscited by 110
  3. 2024Are self-explanations from Large Language Models faithful?cited by 101
  4. 2023Phenomenal Yet Puzzling: Testing Inductive Reasoning Capabilities of Language Models with Hypothesis Refinementcited by 89
  5. 2024DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflowscited by 43
  6. 2024Consistency Checks for Language Model Forecasterscited by 27
  7. 2024Towards Consistent Natural-Language Explanations via Explanation-Consistency Finetuningcited by 26
  8. 2024Deductive Closure Training of Language Models for Coherence, Accuracy, and Updatabilitycited by 22
  9. 2024Unveiling the Tapestry of Consistency in Large Vision-Language Modelscited by 22
  10. 2024A Survey on the Honesty of Large Language Modelscited by 21
  11. 2024BeHonest: Benchmarking Honesty in Large Language Modelscited by 20
  12. 2024Exploring Response Uncertainty in MLLMs: An Empirical Evaluation under Misleading Scenarioscited by 19
31 more
  1. 2024On the Consistency of Video Large Language Models in Temporal Comprehensioncited by 18
  2. 2025Consistency in Language Models: Current Landscape, Challenges, and Future Directionscited by 14
  3. 2025Potemkin Understanding in Large Language Modelscited by 14
  4. 2025Calibrating Verbalized Confidence with Self-Generated Distractorscited by 14
  5. 2025When Continue Learning Meets Multimodal Large Language Model: A Surveycited by 8
  6. 2025Line of Duty: Evaluating LLM Self-Knowledge via Consistency in Feasibility Boundariescited by 7
  7. 2024Retrospective Learning from Interactionscited by 6
  8. 2024DCR: Divide-and-Conquer Reasoning for Multi-choice Question Answering with LLMscited by 5
  9. 2025KnowRL: Teaching Language Models to Know What They Knowcited by 5
  10. 2026Do explanations generalize across large reasoning models?cited by 5
  11. 2024ContraSolver: Self-Alignment of Language Models by Resolving Internal Preference Contradictionscited by 4
  12. 2025Uncovering Gaps in How Humans and LLMs Interpret Subjective Languagecited by 4
  13. 2025RankAlign: A Ranking View of the Generator-Validator Gap in Large Language Modelscited by 4
  14. 2025LM Agents May Fail to Act on Their Own Risk Knowledgecited by 4
  15. 2026Aletheia: What Makes RLVR For Code Verifiers Tick?cited by 4
  16. 2025Beyond RLHF: A Unified Theoretical Framework of Alignmentcited by 3
  17. 2025MedVAL: Toward Expert-Level Medical Text Validation with Language Modelscited by 3
  18. 2025On Evaluating LLM Alignment by Evaluating LLMs as Judgescited by 3
  19. 2024Variability Need Not Imply Error: The Case of Adequate but Semantically Distinct Responsescited by 2
  20. 2025Mirage of Mastery: Memorization Tricks LLMs into Artificially Inflated Self-Knowledgecited by 2
  21. 2026When Reasoning Traces Become Performative: Step-Level Evidence that Chain-of-Thought Is an Imperfect Oversight Channelcited by 2
  22. 2026Partition, Prompt, Aggregate: Statistical Self-Consistency in Language Modelscited by 2
  23. 2025HalluClean: A Unified Framework to Combat Hallucinations in LLMscited by 1
  24. 2026Unlearners Can Lie: Evaluating and Improving Honesty in LLM Unlearningcited by 1
  25. 2026Self-Verified Distillation: Your Language Model Is Secretly Its Own Synthetic Data Pipelinecited by 1
  26. 2026Beyond Calibration: Do a Typed-Decision Model's Probabilities Obey the Probability Axioms?cited by 1
  27. 2023Evaluating, Understanding, and Improving Constrained Text Generation for Large Language Modelscited by 0
  28. 2026The Metanym Game: An LLM Benchmark Without Ground Truth That Rises With the Models It Measurescited by 0
  29. 2026The Consistency Dilemma in LLMs: Generator-Evaluator Agreement and Vulnerability to Mistakescited by 0
  30. 2026Improving LLMs via Validator-to-Generator Alignmentcited by 0
  31. 2026Knowing Is Not Choosing: What Explicit Verification Adds Beyond Generative Preferencecited by 0