mlsanityData to 7 Oct 2026

LLMs as Factual Reasoners: Insights from Existing Benchmarks and Beyond

arXiv 2305.14540 ↗First posted 23 May 2023cs.CLcited by 37

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

With the recent appearance of LLMs in practical settings, having methods that can effectively detect factual inconsistencies is crucial to reduce the propagation of misinformation and improve trust in model outputs. When testing on existing factual consistency benchmarks, we find that a few large language models (LLMs) perform competitively on classification benchmarks for factual inconsistency detection compared to traditional non-LLM methods. However, a closer analysis reveals that most LLMs fail on more complex formulations of the task and exposes issues with existing evaluation benchmarks, affecting evaluation precision. To address this, we propose a new protocol for inconsistency detection benchmark creation and implement it in a 10-domain benchmark called SummEdits. This new benchmark is 20 times more cost-effective per sample than previous benchmarks and highly reproducible, as we estimate inter-annotator agreement at about 0.9. Most LLMs struggle on SummEdits, with performance close to random chance. The best-performing model, GPT-4, is still 8% below estimated human performance, highlighting the gaps in LLMs' ability to reason about facts and detect inconsistencies when they occur.

Key facts. Cited by 37 papers in the map (data through 7 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 17 of its 50 references that are in the map. First posted on arXiv 23 May 2023.

Builds on

17 of 50 references.

Closest ancestors · 2

Same direction, newest first

  1. 2023LLaMA: Open and Efficient Foundation Language Models27 Feb
  2. 2022The Unreliability of Explanations in Few-shot Prompting for Textual Reasoning6 May

Other references · 15

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
  3. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,205
  4. 2018Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarizationcited by 1,308
  5. 2022LaMDA: Language Models for Dialog Applicationscited by 1,218
  6. 2023A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPTcited by 612
  7. 2023GPTScore: Evaluate as You Desirecited by 357
  8. 2022News Summarization and Evaluation in the Era of GPT-3cited by 324
  9. 2022Can language models learn from explanations in context?cited by 248
  10. 2020Evaluating Factuality in Generation with Dependency-level Entailmentcited by 122
  11. 2022Understanding Factual Errors in Summarization: Errors, Summarizers, Datasets, Error Detectorscited by 103
  12. 2022ECTSum: A New Benchmark Dataset For Bullet Point Summarization of Long Earnings Call Transcriptscited by 59
3 more
  1. 2022Multi-LexSum: Real-World Summaries of Civil Rights Lawsuits at Multiple Granularitiescited by 47
  2. 2022PREME: Preference-based Meeting Exploration through an Interactive Questionnairecited by 2
  3. 2020The Spotify Podcast Datasetcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 12; 2024: 15; 2025: 8; 2026: 2.

37 papers.

Sort:

By area:

  1. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,582
  2. 2025Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Modelscited by 291
  3. 2023Art or Artifice? Large Language Models and the False Promise of Creativitycited by 168
  4. 2024Faithfulness vs. Plausibility: On the (Un)Reliability of Explanations from Large Language Modelscited by 104
  5. 2023Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experimentcited by 71
  6. 2023Benchmarking Generation and Evaluation Capabilities of Large Language Models for Instruction Controllable Summarizationcited by 69
  7. 2024Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systemscited by 63
  8. 2024Securing Large Language Models: Threats, Vulnerabilities and Responsible Practicescited by 36
  9. 2023Beyond the Chat: Executable and Verifiable Text-Editing with LLMscited by 35
  10. 2023Foundation Metrics for Evaluating Effectiveness of Healthcare Conversations Powered by Generative AIcited by 32
  11. 2024How Far Are We from Intelligent Visual Deductive Reasoning?cited by 29
  12. 2024Take Caution in Using LLMs as Human Surrogates: Scylla Ex Machinacited by 29
25 more
  1. 2024LeanReasoner: Boosting Complex Logical Reasoning with Leancited by 28
  2. 2023Probing the Multi-turn Planning Capabilities of LLMs via 20 Question Gamescited by 24
  3. 2023Are Large Language Models Reliable Judges? A Study on the Factuality Evaluation Capabilities of LLMscited by 23
  4. 2023ChatGPT's One-year Anniversary: Are Open-Source Large Language Models Catching up?cited by 15
  5. 2025RvLLM: LLM Runtime Verification with Domain Knowledgecited by 15
  6. 2024Evaluating LLM Reasoning in the Operations Research Domain with ORQAcited by 12
  7. 2023AMRFact: Enhancing Summarization Factuality Evaluation with AMR-Driven Negative Samples Generationcited by 10
  8. 2024SQLucid: Grounding Natural Language Database Queries with Interactive Explanationscited by 10
  9. 2025DeepTRACE: Auditing Deep Research AI Systems for Tracking Reliability Across Citations and Evidencecited by 10
  10. 2024ERBench: An Entity-Relationship based Automatically Verifiable Hallucination Benchmark for Large Language Modelscited by 9
  11. 2024Automatic Generation of Model and Data Cards: A Step Towards Responsible AIcited by 9
  12. 2024Rationale-Aware Answer Verification by Pairwise Self-Evaluationcited by 8
  13. 2024Integration of cognitive tasks into artificial general intelligence test for large modelscited by 5
  14. 2024Integrating Large Language Models and Knowledge Graphs for Extraction and Validation of Textual Test Datacited by 5
  15. 2023FaNS: a Facet-based Narrative Similarity Metriccited by 3
  16. 2024Less is More for Improving Automatic Evaluation of Factual Consistencycited by 2
  17. 2025HIDE and Seek: Detecting Hallucinations in Language Models via Decoupled Representationscited by 2
  18. 2023Salespeople vs SalesBot: Exploring the Role of Educational Value in Conversational Recommender Systemscited by 1
  19. 2025Learning Auxiliary Tasks Improves Reference-Free Hallucination Detection in Open-Domain Long-Form Generationcited by 1
  20. 2025Misleading through Inconsistency: A Benchmark for Political Inconsistencies Detectioncited by 1
  21. 2025Beyond Benchmarks: LLM Evaluation with an Anthropomorphic and Lifecycle-oriented Roadmapcited by 1
  22. 2025PrefixNLI: Detecting Factual Inconsistencies as Soon as They Arisecited by 1
  23. 2024LoraMap: Harnessing the Power of LoRA Connectionscited by 0
  24. 2026Probabilistic Guarantees for Reducing Contextual Hallucinations in LLMscited by 0
  25. 2026Resonant Context Anchoring: Decoupling Attention Routing and Signal Gain at Inference Timecited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026InFi-Check: Interpretable and Fine-Grained Fact-Checking of LLMs
  2. 2024Evaluating Consistency and Reasoning Capabilities of Large Language Models
  3. 2024Logical Consistency of Large Language Models in Fact-checking
  4. 2026Can LLM Reasoning Be Trusted? A Comparative Study: Using Human Benchmarking on Statistical Tasks
  5. 2023Evaluating Factual Consistency of Summaries with Large Language Models
  6. 2026Towards Comprehensive Stage-wise Benchmarking of Large Language Models in Fact-Checking
  7. 2025Is LLM an Overconfident Judge? Unveiling the Capabilities of LLMs in Detecting Offensive Language with Annotation Disagreement
  8. 2024Evaluating LLMs at Detecting Errors in LLM Responses