mlsanityData to 14 Oct 2026

FRUIT: Faithfully Reflecting Updated Information in Text

arXiv 2112.08634 ↗First posted 16 Dec 2021cs.CLcited by 39

Not on the map. Why?

Abstract (arXiv)

Textual knowledge bases such as Wikipedia require considerable effort to keep up to date and consistent. While automated writing assistants could potentially ease this burden, the problem of suggesting edits grounded in external knowledge has been under-explored. In this paper, we introduce the novel generation task of *faithfully reflecting updated information in text* (FRUIT) where the goal is to update an existing article given new evidence. We release the FRUIT-WIKI dataset, a collection of over 170K distantly supervised data produced from pairs of Wikipedia snapshots, along with our data generation pipeline and a gold evaluation set of 914 instances whose edits are guaranteed to be supported by the evidence. We provide benchmark results for popular generation systems as well as EDIT5 -- a T5-based approach tailored to editing we introduce that establishes the state of the art. Our analysis shows that developing models that can update articles faithfully requires new capabilities for neural generation models, and opens doors to many new applications.

Key facts. Cited by 39 papers in the map (data through 14 Oct 2026). Builds on 24 of its 41 references that are in the map. First posted on arXiv 16 Dec 2021.

Builds on

24 of 41 references.

References · 24

  1. 2017Attention Is All You Needcited by 60,361
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  3. 2014Sequence to Sequence Learning with Neural Networkscited by 2,986
  4. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,141
  5. 2015Microsoft COCO Captions: Data Collection and Evaluation Servercited by 1,462
  6. 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,375
  7. 2017Get To The Point: Summarization with Pointer-Generator Networkscited by 1,292
  8. 2018Adafactor: Adaptive Learning Rates with Sublinear Memory Costcited by 905
  9. 2019CTRL: A Conditional Transformer Language Model for Controllable Generationcited by 762
  10. 2021Entity-Based Knowledge Conflicts in Question Answeringcited by 356
  11. 2018Generating Wikipedia by Summarizing Long Sequencescited by 335
  12. 2021Time-Aware Language Models as Temporal Knowledge Basescited by 296
12 more
  1. 2020Evaluation of Text Generation: A Surveycited by 277
  2. 2020ToTTo: A Controlled Table-To-Text Generation Datasetcited by 268
  3. 2017Challenges in Data-to-Document Generationcited by 226
  4. 2021Get Your Vitamin C! Robust Fact Verification with Contrastive Evidencecited by 225
  5. 2021Hurdles to Progress in Long-form Question Answeringcited by 167
  6. 2019Gmail Smart Compose: Real-Time Assisted Writingcited by 98
  7. 2018Table-to-Text: Describing Table Region with Natural Languagecited by 41
  8. 2020WikiTableT: A Large-Scale Data-to-Text Dataset for Generating Wikipedia Article Sectionscited by 22
  9. 2020A Comprehensive Survey of Grammar Error Correctioncited by 17
  10. 2021IGA : An Intent-Guided Authoring Assistantcited by 12
  11. 2019PoMo: Generating Entity-Specific Post-Modifiers in Contextcited by 4
  12. 2021DESCGEN: A Distantly Supervised Dataset for Generating Abstractive Entity Descriptionscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 8; 2023: 16; 2024: 8; 2025: 5; 2026: 2.

39 papers.

Sort:

By area:

  1. 2022RARR: Researching and Revising What Language Models Say, Using Language Modelscited by 287
  2. 2024When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMscited by 240
  3. 2023Can Pre-trained Vision and Language Models Answer Visual Information-Seeking Questions?cited by 207
  4. 2023Knowledge Editing for Large Language Models: A Surveycited by 194
  5. 2022TemporalWiki: A Lifelong Benchmark for Training and Evaluating Ever-Evolving Language Modelscited by 119
  6. 2023Factcheck-Bench: Fine-Grained Evaluation Benchmark for Automatic Fact-checkerscited by 87
  7. 2022PEER: A Collaborative Language Modelcited by 86
  8. 2024A Design Space for Intelligent and Interactive Writing Assistantscited by 76
  9. 2023RewriteLM: An Instruction-Tuned Large Language Model for Text Rewritingcited by 67
  10. 2023SummIt: Iterative Text Summarization via ChatGPTcited by 54
  11. 2023PURR: Efficiently Editing Language Model Hallucinations by Denoising Language Model Corruptionscited by 53
  12. 2022CoditT5: Pretraining for Source Code and Natural Language Editingcited by 45
27 more
  1. 2022Stretching Sentence-pair NLI Models to Reason over Long Documents and Clusterscited by 44
  2. 2023Self-Checker: Plug-and-Play Modules for Fact-Checking with Large Language Modelscited by 31
  3. 2022EditEval: An Instruction-Based Benchmark for Text Improvementscited by 30
  4. 2023WebBrain: Learning to Generate Factually Correct Articles for Queries by Grounding on Large Web Corpuscited by 29
  5. 2023TempTabQA: Temporal Question Answering for Semi-Structured Tablescited by 27
  6. 2024Nearest Neighbor Speculative Decoding for LLM Generation and Attributioncited by 25
  7. 2022OASum: Large-Scale Open Domain Aspect-based Summarizationcited by 24
  8. 2024Enhancing Temporal Understanding in LLMs for Semi-structured Tablescited by 8
  9. 2023Towards an On-device Agent for Text Rewritingcited by 7
  10. 2023XATU: A Fine-grained Instruction-based Benchmark for Explainable Text Updatescited by 6
  11. 2024MisgenderMender: A Community-Informed Approach to Interventions for Misgenderingcited by 6
  12. 2025DocMEdit: Towards Document-Level Model Editingcited by 6
  13. 2023Fast, Scalable, Warm-Start Semidefinite Programming with Spectral Bundling and Sketchingcited by 4
  14. 2024Proofread: Fixes All Errors with One Tapcited by 4
  15. 2023DIFFQG: Generating Questions to Summarize Factual Changescited by 3
  16. 2023InfoSync: Information Synchronization across Multilingual Semi-structured Tablescited by 3
  17. 2024Medico: Towards Hallucination Detection and Correction with Multi-source Evidence Fusioncited by 3
  18. 2025REVERSUM: A Multi-staged Retrieval-Augmented Generation Method to Enhance Wikipedia Tail Biographies through Personal Narrativescited by 3
  19. 2025Dr Genre: Reinforcement Learning from Decoupled LLM Feedback for Generic Text Rewritingcited by 3
  20. 2023AWESOME: GPU Memory-constrained Long Document Summarization using Memory Mechanism and Global Salient Contentcited by 2
  21. 2024WIKIGENBENCH: Exploring Full-length Wikipedia Generation under Real-World Scenariocited by 2
  22. 2025Leveraging LLM For Synchronizing Information Across Multilingual Tablescited by 2
  23. 2023From Nuisance to News Sense: Augmenting the News with Cross-Document Evidence and Contextcited by 1
  24. 2022An Inclusive Notion of Textcited by 0
  25. 2025Excision Score: Evaluating Edits with Surgical Precisioncited by 0
  26. 2026Intention-Adaptive LLM Fine-Tuning for Text Revision Generationcited by 0
  27. 2026Knowledge Pull Requests for Continual Document Authoringcited by 0