mlsanityData to 7 Oct 2026

Molecular Facts: Desiderata for Decontextualization in LLM Fact Verification

arXiv 2406.20079 ↗First posted 28 Jun 2024cs.CL cs.AIcited by 42

Not on the map. Why?

Abstract (arXiv)

Automatic factuality verification of large language model (LLM) generations is becoming more and more widely used to combat hallucinations. A major point of tension in the literature is the granularity of this fact-checking: larger chunks of text are hard to fact-check, but more atomic facts like propositions may lack context to interpret correctly. In this work, we assess the role of context in these atomic facts. We argue that fully atomic facts are not the right representation, and define two criteria for molecular facts: decontextuality, or how well they can stand alone, and minimality, or how little extra information is added to achieve decontexuality. We quantify the impact of decontextualization on minimality, then present a baseline methodology for generating molecular facts automatically, aiming to add the right amount of information. We compare against various methods of decontextualization and find that molecular facts balance minimality with fact verification accuracy in ambiguous settings.

Key facts. Cited by 42 papers in the map (data through 7 Oct 2026). Builds on 27 of its 47 references that are in the map. First posted on arXiv 28 Jun 2024.

Builds on

27 of 47 references.

References · 27

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,259
  3. 2023Fine-Grained Human Feedback Gives Better Rewards for Language Model Trainingcited by 399
  4. 2023Evaluating Verifiability in Generative Search Enginescited by 297
  5. 2023AlignScore: Evaluating Factual Consistency with a Unified Alignment Functioncited by 284
  6. 2021Measuring Attribution in Natural Language Generation Modelscited by 253
  7. 2024MiniCheck: Efficient Fact-Checking of LLMs on Grounding Documentscited by 251
  8. 2021QAFactEval: Improved QA-Based Factual Consistency Evaluation for Summarizationcited by 233
  9. 2022WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creationcited by 200
  10. 2021SituatedQA: Incorporating Extra-Linguistic Contexts into QAcited by 145
  11. 2022Rethinking with Retrieval: Faithful Large Language Model Inferencecited by 145
  12. 2022Revisiting the Gold Standard: Grounding Summarization Evaluation with Robust Human Evaluationcited by 141
15 more
  1. 2021Annotating and Modeling Fine-grained Factuality in Summarizationcited by 131
  2. 2023WiCE: Real-World Entailment for Claims in Wikipediacited by 118
  3. 2023Complex Claim Verification with Evidence Retrieved in the Wildcited by 110
  4. 2021Decontextualization: Making Sentences Stand-Alonecited by 101
  5. 2023Factcheck-Bench: Fine-Grained Evaluation Benchmark for Automatic Fact-checkerscited by 86
  6. 2023Factually Consistent Summarization via Reinforcement Learning with Textual Entailment Feedbackcited by 71
  7. 2021Finding a Balanced Degree of Automation for Summary Evaluationcited by 43
  8. 2024A Closer Look at Claim Decompositioncited by 40
  9. 2022PropSegmEnt: A Large-Scale Corpus for Proposition-Level Segmentation and Entailment Recognitioncited by 29
  10. 2024AmbigDocs: Reasoning across Documents on Different Entities under the Same Namecited by 21
  11. 2022Shortcomings of Question Answering Based Factuality Frameworks for Error Localizationcited by 15
  12. 2023A Question Answering Framework for Decontextualizing User-facing Snippets from Scientific Documentscited by 15
  13. 2024Merging Facts, Crafting Fallacies: Evaluating the Contradictory Nature of Aggregated Factual Claims in Long-Form Generationscited by 14
  14. 2023Concise Answers to Complex Questions: Summarization of Long-form Answerscited by 7
  15. 2023Drafting Event Schemas using Language Modelscited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 6; 2025: 23; 2026: 13.

42 papers.

Sort:

By area:

  1. 2024Decomposition Dilemmas: Does Claim Decomposition Boost or Burden Fact-Checking Performance?cited by 34
  2. 2024FactBench: A Dynamic Benchmark for In-the-Wild Language Model Factuality Evaluationcited by 23
  3. 2024DnDScore: Decontextualization and Decomposition for Factuality Verification in Long-Form Text Generationcited by 22
  4. 2025Conformal Linguistic Calibration: Trading-off between Factuality and Specificitycited by 20
  5. 2025Towards Effective Extraction and Evaluation of Factual Claimscited by 18
  6. 2025WikiVideo: Article Generation from Multiple Videoscited by 13
  7. 2025LAQuer: Localized Attribution Queries in Content-grounded Generationcited by 13
  8. 2025VeriFact: Enhancing Long-Form Factuality Evaluation with Refined Fact Extraction and Reference Factscited by 11
  9. 2024QAPyramid: Fine-grained Evaluation of Content Selection for Text Summarizationcited by 8
  10. 2025A New HOPE: Domain-agnostic Automatic Evaluation of Text Chunkingcited by 7
  11. 2025BottleHumor: Self-Informed Humor Explanation using the Information Bottleneck Principlecited by 6
  12. 2025All Claims Are Equal, but Some Claims Are More Equal Than Others: Importance-Sensitive Factuality Evaluation of LLM Generationscited by 6
30 more
  1. 2024FactEHR: A Dataset for Evaluating Factuality in Clinical Notes Using LLMscited by 5
  2. 2025GenerationPrograms: Fine-grained Attribution with Executable Programscited by 4
  3. 2025Detecting Corpus-Level Knowledge Inconsistencies in Wikipedia with Large Language Modelscited by 4
  4. 2025BOOST: Bootstrapping Strategy-Driven Reasoning Programs for Program-Guided Fact-Checkingcited by 3
  5. 2025The Curious Case of Factuality Finetuning: Models' Internal Beliefs Can Improve Factualitycited by 3
  6. 2025Context-Efficient Retrieval with Factual Decompositioncited by 2
  7. 2025PASS-FC: Progressive and Adaptive Search Scheme for Fact Checking of Comprehensive Claimscited by 2
  8. 2025SUCEA: Reasoning-Intensive Retrieval for Adversarial Fact-checking through Claim Decomposition and Editingcited by 2
  9. 2025Fact in Fragments: Deconstructing Complex Claims via LLM-based Atomic Fact Extraction and Verificationcited by 2
  10. 2025Decide less, communicate more: On the construct validity of end-to-end fact-checking in medicinecited by 2
  11. 2025Chronological Passage Assembling in RAG framework for Temporal Question Answeringcited by 2
  12. 2025T-FIX: Text-Based Explanations with Features Interpretable to eXpertscited by 2
  13. 2026Distill and Align Decomposition for Enhanced Claim Verificationcited by 2
  14. 2026Attributable by Construction: Claim-Anchored Provenance for Multi-Document Summarizationcited by 2
  15. 2026When Summaries Distort Decisions: Information Fidelity in LLM-Compressed Financial Analysiscited by 2
  16. 2025Exploration of Plan-Guided Summarization for Narrative Texts: the Case of Small Language Modelscited by 1
  17. 2025Seeing Through the MiRAGE: Evaluating Multimodal Retrieval Augmented Generationcited by 1
  18. 2026IRB: Automated Generation of Robust Factuality Benchmarkscited by 1
  19. 2026From Articles to Premises: Building PrimeFacts, an Extraction Methodology and Resource for Fact-Checking Evidencecited by 1
  20. 2024Doppelg"anger's Watch: A Split Objective Approach to Large Language Modelscited by 0
  21. 2025How Grounded is Wikipedia? A Study on Structured Evidential Support and Retrievalcited by 0
  22. 2025HiFACTMix: A Code-Mixed Benchmark and Graph-Aware Model for EvidenceBased Political Claim Verification in Hinglishcited by 0
  23. 2026eTracer: Polarity-Aware Evidence Groundingcited by 0
  24. 2026The Warrant Gap: Claim-Conditioned Re-scoring for Fact-Checkingcited by 0
  25. 2026VetScore: Risk-Weighted Fact Verification for Veterinary Long-Form QA with Citationscited by 0
  26. 2026TriQua: Reconciling Granularity and Context in Factuality Evaluationcited by 0
  27. 2026Decomposed Entailment for Factuality Checking and Hallucination Detectioncited by 0
  28. 2026MedSNIP: Building and Benchmarking Snippet-Level Granularity for Medical Fact Verificationcited by 0
  29. 2026Knowledge Pull Requests for Continual Document Authoringcited by 0
  30. 2026MultiVENT-Raw: A Benchmark for Retrieval and Reasoning over Raw Videoscited by 0