mlsanityData to 14 Oct 2026

Converge to the Truth: Factual Error Correction via Iterative Constrained Editing

arXiv 2211.12130 ↗First posted 22 Nov 2022cs.CLcited by 9

Not on the map. Why?

Abstract (arXiv)

Given a possibly false claim sentence, how can we automatically correct it with minimal editing? Existing methods either require a large number of pairs of false and corrected claims for supervised training or do not handle well errors spanning over multiple tokens within an utterance. In this paper, we propose VENCE, a novel method for factual error correction (FEC) with minimal edits. VENCE formulates the FEC problem as iterative sampling editing actions with respect to a target density function. We carefully design the target function with predicted truthfulness scores from an offline trained fact verification model. VENCE samples the most probable editing positions based on back-calculated gradients of the truthfulness score concerning input tokens and the editing actions using a distantly-supervised language model (T5). Experiments on a public dataset show that VENCE improves the well-adopted SARI metric by 5.3 (or a relative improvement of 11.8%) over the previous best distantly-supervised methods.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 36 of its 57 references that are in the map. First posted on arXiv 22 Nov 2022.

Builds on

36 of 57 references.

References · 36

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  3. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  4. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  5. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,893
  6. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  7. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
  8. 2017A Broad-Coverage Challenge Corpus for Sentence Understanding through Inferencecited by 2,682
  9. 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,375
  10. 2020Fact or Fiction: Verifying Scientific Claimscited by 554
  11. 2020Autoregressive Entity Retrievalcited by 334
  12. 2019Levenshtein Transformercited by 199
24 more
  1. 2020GECToR -- Grammatical Error Correction: Tag, Not Rewritecited by 177
  2. 2021On Hallucination and Predictive Uncertainty in Conditional Language Generationcited by 176
  3. 2019MultiFC: A Real-World Multi-Domain Dataset for Evidence-Based Fact Checking of Claimscited by 169
  4. 2021The Curious Case of Hallucinations in Neural Machine Translationcited by 161
  5. 2017Faithful to the Original: Fact Aware Neural Abstractive Summarizationcited by 141
  6. 2021Q2: Evaluating Factual Consistency in Knowledge-Grounded Dialogues via Question Generation and Question Answeringcited by 132
  7. 2021Hallucinated but Factual! Inspecting the Factuality of Hallucinations in Abstractive Summarizationcited by 129
  8. 2018CGMH: Constrained Sentence Generation by Metropolis-Hastings Samplingcited by 94
  9. 2020Explainable Automated Fact-Checking: A Surveycited by 83
  10. 2022Mix and Match: Learning-free Controllable Text Generation using Energy Language Modelscited by 66
  11. 2021ProoFVer: Natural Logic Theorem Proving for Fact Verificationcited by 53
  12. 2020Evidence-based Factual Error Correctioncited by 44
  13. 2021Exposing the Implicit Energy Networks behind Masked Language Models via Metropolis--Hastingscited by 37
  14. 2018A Multilayer Convolutional Encoder-Decoder Neural Network for Grammatical Error Correctioncited by 36
  15. 2022EdiT5: Semi-Autoregressive Text-Editing with T5 Warm-Startcited by 35
  16. 2022Evaluating Factuality in Text Simplificationcited by 34
  17. 2021LM-Critic: Language Models for Unsupervised Grammatical Error Correctioncited by 33
  18. 2019Automatic Fact-guided Sentence Modificationcited by 28
  19. 2022An Imitation Learning Curriculum for Text Editing with Non-Autoregressive Modelscited by 20
  20. 2020LOREN: Logic-Regularized Reasoning for Interpretable Fact Verificationcited by 16
  21. 2020Fact-based Text Editingcited by 14
  22. 2020Data-to-Text Generation with Iterative Text Editingcited by 11
  23. 2021Unified Dual-view Cognitive Model for Interpretable Claim Verificationcited by 10
  24. 2021Unsupervised Editing for Counterfactual Storiescited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 4; 2024: 1; 2025: 1; 2026: 3.

9 papers.

  1. 2023Dancing Between Success and Failure: Edit-level Simplification Evaluation using SALSAcited by 15
  2. 2023Reference Matters: Benchmarking Factual Error Correction for Dialogue Summarization with Fine-grained Evaluation Frameworkcited by 7
  3. 2024Zero-Shot Fact Verification via Natural Logic and Large Language Modelscited by 5
  4. 2023Improving Factual Error Correction by Learning to Inject Factual Errorscited by 4
  5. 2023SciFix: Outperforming GPT3 on Scientific Factual Error Correctioncited by 1
  6. 2025HAVE: Head-Adaptive Gating and ValuE Calibration for Hallucination Mitigation in Large Language Modelscited by 0
  7. 2026CECOR: Correction-oriented synthetic data construction for factual error correctioncited by 0
  8. 2026Mask-to-Correct^+: Leveraging Retriever Diversity for Masking-guided Faithful Fact Correctioncited by 0
  9. 2026An Empirical Analysis of Factual Errors in Human-Written Text and Its Application to Factual Error Detectioncited by 0