# Molecular Facts: Desiderata for Decontextualization in LLM Fact Verification

arXiv: https://arxiv.org/abs/2406.20079 · HTML: https://mlsanity.com/paper/2406.20079/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 47 references that are in the map.
- First posted on arXiv 28 Jun 2024.

## Builds on (most cited first, 27 of 27)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation](https://mlsanity.com/paper/2305.14251/) (2023)
- [Fine-Grained Human Feedback Gives Better Rewards for Language Model Training](https://mlsanity.com/paper/2306.01693/) (2023)
- [Evaluating Verifiability in Generative Search Engines](https://mlsanity.com/paper/2304.09848/) (2023)
- [AlignScore: Evaluating Factual Consistency with a Unified Alignment Function](https://mlsanity.com/paper/2305.16739/) (2023)
- [MiniCheck: Efficient Fact-Checking of LLMs on Grounding Documents](https://mlsanity.com/paper/2404.10774/) (2024)
- [Measuring Attribution in Natural Language Generation Models](https://mlsanity.com/paper/2112.12870/) (2021)
- [QAFactEval: Improved QA-Based Factual Consistency Evaluation for Summarization](https://mlsanity.com/paper/2112.08542/) (2021)
- [WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creation](https://mlsanity.com/paper/2201.05955/) (2022)
- [SituatedQA: Incorporating Extra-Linguistic Contexts into QA](https://mlsanity.com/paper/2109.06157/) (2021)
- [Rethinking with Retrieval: Faithful Large Language Model Inference](https://mlsanity.com/paper/2301.00303/) (2022)
- [Revisiting the Gold Standard: Grounding Summarization Evaluation with Robust Human Evaluation](https://mlsanity.com/paper/2212.07981/) (2022)
- [Annotating and Modeling Fine-grained Factuality in Summarization](https://mlsanity.com/paper/2104.04302/) (2021)
- [WiCE: Real-World Entailment for Claims in Wikipedia](https://mlsanity.com/paper/2303.01432/) (2023)
- [Complex Claim Verification with Evidence Retrieved in the Wild](https://mlsanity.com/paper/2305.11859/) (2023)
- [Decontextualization: Making Sentences Stand-Alone](https://mlsanity.com/paper/2102.05169/) (2021)
- [Factcheck-Bench: Fine-Grained Evaluation Benchmark for Automatic Fact-checkers](https://mlsanity.com/paper/2311.09000/) (2023)
- [Factually Consistent Summarization via Reinforcement Learning with Textual Entailment Feedback](https://mlsanity.com/paper/2306.00186/) (2023)
- [Finding a Balanced Degree of Automation for Summary Evaluation](https://mlsanity.com/paper/2109.11503/) (2021)
- [A Closer Look at Claim Decomposition](https://mlsanity.com/paper/2403.11903/) (2024)
- [PropSegmEnt: A Large-Scale Corpus for Proposition-Level Segmentation and Entailment Recognition](https://mlsanity.com/paper/2212.10750/) (2022)
- [AmbigDocs: Reasoning across Documents on Different Entities under the Same Name](https://mlsanity.com/paper/2404.12447/) (2024)
- [A Question Answering Framework for Decontextualizing User-facing Snippets from Scientific Documents](https://mlsanity.com/paper/2305.14772/) (2023)
- [Shortcomings of Question Answering Based Factuality Frameworks for Error Localization](https://mlsanity.com/paper/2210.06748/) (2022)
- [Merging Facts, Crafting Fallacies: Evaluating the Contradictory Nature of Aggregated Factual Claims in Long-Form Generations](https://mlsanity.com/paper/2402.05629/) (2024)
- [Concise Answers to Complex Questions: Summarization of Long-form Answers](https://mlsanity.com/paper/2305.19271/) (2023)
- [Drafting Event Schemas using Language Models](https://mlsanity.com/paper/2305.14847/) (2023)

## Built on by (most cited first, 30 of 44)

- [Decomposition Dilemmas: Does Claim Decomposition Boost or Burden Fact-Checking Performance?](https://mlsanity.com/paper/2411.02400/) (2024)
- [FactBench: A Dynamic Benchmark for In-the-Wild Language Model Factuality Evaluation](https://mlsanity.com/paper/2410.22257/) (2024)
- [DnDScore: Decontextualization and Decomposition for Factuality Verification in Long-Form Text Generation](https://mlsanity.com/paper/2412.13175/) (2024)
- [Conformal Linguistic Calibration: Trading-off between Factuality and Specificity](https://mlsanity.com/paper/2502.19110/) (2025)
- [Towards Effective Extraction and Evaluation of Factual Claims](https://mlsanity.com/paper/2502.10855/) (2025)
- [WikiVideo: Article Generation from Multiple Videos](https://mlsanity.com/paper/2504.00939/) (2025)
- [LAQuer: Localized Attribution Queries in Content-grounded Generation](https://mlsanity.com/paper/2506.01187/) (2025)
- [VeriFact: Enhancing Long-Form Factuality Evaluation with Refined Fact Extraction and Reference Facts](https://mlsanity.com/paper/2505.09701/) (2025)
- [QAPyramid: Fine-grained Evaluation of Content Selection for Text Summarization](https://mlsanity.com/paper/2412.07096/) (2024)
- [A New HOPE: Domain-agnostic Automatic Evaluation of Text Chunking](https://mlsanity.com/paper/2505.02171/) (2025)
- [BottleHumor: Self-Informed Humor Explanation using the Information Bottleneck Principle](https://mlsanity.com/paper/2502.18331/) (2025)
- [All Claims Are Equal, but Some Claims Are More Equal Than Others: Importance-Sensitive Factuality Evaluation of LLM Generations](https://mlsanity.com/paper/2510.07083/) (2025)
- [FactEHR: A Dataset for Evaluating Factuality in Clinical Notes Using LLMs](https://mlsanity.com/paper/2412.12422/) (2024)
- [GenerationPrograms: Fine-grained Attribution with Executable Programs](https://mlsanity.com/paper/2506.14580/) (2025)
- [The Curious Case of Factuality Finetuning: Models' Internal Beliefs Can Improve Factuality](https://mlsanity.com/paper/2507.08371/) (2025)
- [Detecting Corpus-Level Knowledge Inconsistencies in Wikipedia with Large Language Models](https://mlsanity.com/paper/2509.23233/) (2025)
- [BOOST: Bootstrapping Strategy-Driven Reasoning Programs for Program-Guided Fact-Checking](https://mlsanity.com/paper/2504.02467/) (2025)
- [Context-Efficient Retrieval with Factual Decomposition](https://mlsanity.com/paper/2503.19574/) (2025)
- [PASS-FC: Progressive and Adaptive Search Scheme for Fact Checking of Comprehensive Claims](https://mlsanity.com/paper/2504.09866/) (2025)
- [SUCEA: Reasoning-Intensive Retrieval for Adversarial Fact-checking through Claim Decomposition and Editing](https://mlsanity.com/paper/2506.04583/) (2025)
- [Fact in Fragments: Deconstructing Complex Claims via LLM-based Atomic Fact Extraction and Verification](https://mlsanity.com/paper/2506.07446/) (2025)
- [Decide less, communicate more: On the construct validity of end-to-end fact-checking in medicine](https://mlsanity.com/paper/2506.20876/) (2025)
- [Chronological Passage Assembling in RAG framework for Temporal Question Answering](https://mlsanity.com/paper/2508.18748/) (2025)
- [T-FIX: Text-Based Explanations with Features Interpretable to eXperts](https://mlsanity.com/paper/2511.04070/) (2025)
- [Distill and Align Decomposition for Enhanced Claim Verification](https://mlsanity.com/paper/2602.21857/) (2026)
- [Attributable by Construction: Claim-Anchored Provenance for Multi-Document Summarization](https://mlsanity.com/paper/2606.23989/) (2026)
- [When Summaries Distort Decisions: Information Fidelity in LLM-Compressed Financial Analysis](https://mlsanity.com/paper/2606.29251/) (2026)
- [Exploration of Plan-Guided Summarization for Narrative Texts: the Case of Small Language Models](https://mlsanity.com/paper/2504.09071/) (2025)
- [Seeing Through the MiRAGE: Evaluating Multimodal Retrieval Augmented Generation](https://mlsanity.com/paper/2510.24870/) (2025)
- [IRB: Automated Generation of Robust Factuality Benchmarks](https://mlsanity.com/paper/2602.08070/) (2026)

## 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.