mlsanityData to 7 Oct 2026

Zero-shot Faithful Factual Error Correction

arXiv 2305.07982 ↗First posted 13 May 2023cs.CLcited by 26

Not on the map. Why?

Abstract (arXiv)

Faithfully correcting factual errors is critical for maintaining the integrity of textual knowledge bases and preventing hallucinations in sequence-to-sequence models. Drawing on humans' ability to identify and correct factual errors, we present a zero-shot framework that formulates questions about input claims, looks for correct answers in the given evidence, and assesses the faithfulness of each correction based on its consistency with the evidence. Our zero-shot framework outperforms fully-supervised approaches, as demonstrated by experiments on the FEVER and SciFact datasets, where our outputs are shown to be more faithful. More importantly, the decomposability nature of our framework inherently provides interpretability. Additionally, to reveal the most suitable metrics for evaluating factual error corrections, we analyze the correlation between commonly used metrics with human judgments in terms of three different dimensions regarding intelligibility and faithfulness.

Key facts. Cited by 26 papers in the map (data through 7 Oct 2026). Builds on 41 of its 64 references that are in the map. First posted on arXiv 13 May 2023.

Builds on

41 of 64 references.

References · 41

  1. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  2. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  3. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,115
  4. 2017A Broad-Coverage Challenge Corpus for Sentence Understanding through Inferencecited by 2,665
  5. 2019BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questionscited by 2,140
  6. 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,362
  7. 2019PubMedQA: A Dataset for Biomedical Research Question Answeringcited by 1,271
  8. 2020On Faithfulness and Factuality in Abstractive Summarizationcited by 895
  9. 2021BARTScore: Evaluating Generated Text as Text Generationcited by 719
  10. 2019Evaluating the Factual Consistency of Abstractive Text Summarizationcited by 578
  11. 2020Fact or Fiction: Verifying Scientific Claimscited by 546
  12. 2020UnifiedQA: Crossing Format Boundaries With a Single QA Systemcited by 475
29 more
  1. 2021SummaC: Re-Visiting NLI-based Models for Inconsistency Detection in Summarizationcited by 402
  2. 2020Asking and Answering Questions to Evaluate the Factual Consistency of Summariescited by 376
  3. 2018FakeNewsNet: A Data Repository with News Content, Social Context and Spatialtemporal Information for Studying Fake News on Social Mediacited by 281
  4. 2021QAFactEval: Improved QA-Based Factual Consistency Evaluation for Summarizationcited by 233
  5. 2019Probing Neural Network Comprehension of Natural Language Argumentscited by 212
  6. 2021CLIFF: Contrastive Learning for Improving Faithfulness and Factuality in Abstractive Summarizationcited by 120
  7. 2020Factual Error Correction for Abstractive Summarization Modelscited by 108
  8. 2018Transforming Question Answering Datasets Into Natural Language Inference Datasetscited by 99
  9. 2021X-FACT: A New Benchmark Dataset for Multilingual Fact Checkingcited by 90
  10. 2020Towards Question-Answering as an Automatic Metric for Evaluating the Content Quality of a Summarycited by 87
  11. 2022PEER: A Collaborative Language Modelcited by 86
  12. 2021DocNLI: A Large-scale Dataset for Document-level Natural Language Inferencecited by 75
  13. 2022Generating Literal and Implied Subquestions to Fact-check Complex Claimscited by 66
  14. 2022Generating Scientific Claims for Zero-Shot Scientific Fact Checkingcited by 59
  15. 2022UnifiedQA-v2: Stronger Generalization via Broader Cross-Format Trainingcited by 57
  16. 2021Improving Factual Consistency of Abstractive Summarization via Question Answeringcited by 55
  17. 2022FactPEGASUS: Factuality-Aware Pre-training and Fine-tuning for Abstractive Summarizationcited by 49
  18. 2020Evidence-based Factual Error Correctioncited by 44
  19. 2021CONFIT: Toward Faithful Dialogue Summarization with Linguistically-Informed Contrastive Fine-tuningcited by 44
  20. 2021Zero-shot Fact Verification by Claim Generationcited by 42
  21. 2022Faking Fake News for Real Fake News Detection: Propaganda-loaded Training Data Generationcited by 41
  22. 2022FactGraph: Evaluating Factuality in Summarization with Semantic Graph Representationscited by 41
  23. 2019Automatic Fact-guided Sentence Modificationcited by 28
  24. 2021MixQG: Neural Question Generation with Mixed Answer Typescited by 28
  25. 2022Learning to Revise References for Faithful Summarizationcited by 28
  26. 2022Crowdsourced Fact-Checking at Twitter: How Does the Crowd Compare With Experts?cited by 16
  27. 2022CONCRETE: Improving Cross-lingual Fact-checking with Cross-lingual Retrievalcited by 14
  28. 2022Improving Factual Consistency in Summarization with Compression-Based Post-Editingcited by 13
  29. 2023SWING: Balancing Coverage and Faithfulness for Dialogue Summarizationcited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 15; 2024: 8; 2025: 1; 2026: 2.

26 papers.

Sort:

By area:

  1. 2023Woodpecker: Hallucination Correction for Multimodal Large Language Modelscited by 223
  2. 2023Factuality Challenges in the Era of Large Language Modelscited by 125
  3. 2023Cognitive Mirage: A Review of Hallucinations in Large Language Modelscited by 83
  4. 2024Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systemscited by 63
  5. 2024From Pixels to Insights: A Survey on Automatic Chart Understanding in the Era of Large Foundation Modelscited by 52
  6. 2024Can LLMs Produce Faithful Explanations For Fact-checking? Towards Faithful Explainable Fact-Checking via Multi-Agent Debatecited by 48
  7. 2023Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioningcited by 47
  8. 2023Self-Checker: Plug-and-Play Modules for Fact-Checking with Large Language Modelscited by 31
  9. 2023Gender Biases in Automatic Evaluation Metrics for Image Captioningcited by 24
  10. 2023Leveraging LLMs for Synthesizing Training Data Across Many Languages in Multilingual Dense Retrievalcited by 24
  11. 2023Zero-Resource Hallucination Prevention for Large Language Modelscited by 19
  12. 2023Decoding the Silent Majority: Inducing Belief Augmented Social Graph with Large Language Model for Response Forecastingcited by 11
14 more
  1. 2023AMRFact: Enhancing Summarization Factuality Evaluation with AMR-Driven Negative Samples Generationcited by 10
  2. 2024SafeWorld: Geo-Diverse Safety Alignmentcited by 10
  3. 2024A Survey of Automatic Hallucination Evaluation on Natural Language Generationcited by 7
  4. 2023Interpretable Automatic Fine-grained Inconsistency Detection in Text Summarizationcited by 6
  5. 2024Diver: Large Language Model Decoding with Span-Level Mutual Information Verificationcited by 5
  6. 2023Towards CausalGPT: A Multi-Agent Approach for Faithful Knowledge Reasoning via Promoting Causal Consistency in LLMscited by 4
  7. 2023Factored Verification: Detecting and Reducing Hallucination in Summaries of Academic Paperscited by 4
  8. 2023Defining a New NLP Playgroundcited by 4
  9. 2025CutPaste&Find: Efficient Multimodal Hallucination Detector with Visual-aid Knowledge Basecited by 4
  10. 2026ChartFI: Benchmarking Faithfulness and Insightfulness of Chart Descriptions from Multimodal Large Language Modelscited by 2
  11. 2023SciFix: Outperforming GPT3 on Scientific Factual Error Correctioncited by 1
  12. 2024ACCORD: Closing the Commonsense Measurability Gapcited by 1
  13. 2024Detecting Response Generation Not Requiring Factual Judgmentcited by 0
  14. 2026Mask-to-Correct^+: Leveraging Retriever Diversity for Masking-guided Faithful Fact Correctioncited by 0