mlsanityData to 14 Oct 2026

Factored Verification: Detecting and Reducing Hallucination in Summaries of Academic Papers

arXiv 2310.10627 ↗First posted 16 Oct 2023cs.CL cs.AIcited by 4

Not on the map. Why?

Abstract (arXiv)

Hallucination plagues even frontier LLMs--but how bad is it really for summarizing academic papers? We evaluate Factored Verification, a simple automated method for detecting hallucinations in abstractive summaries. This method sets a new SotA on hallucination detection in the summarization task of the HaluEval benchmark, achieving 76.2% accuracy. We then use this method to estimate how often language models hallucinate when summarizing across multiple academic papers and find 0.62 hallucinations in the average ChatGPT (16k) summary, 0.84 for GPT-4, and 1.55 for Claude 2. We ask models to self-correct using Factored Critiques and find that this lowers the number of hallucinations to 0.49 for ChatGPT, 0.46 for GPT-4, and 0.95 for Claude 2. The hallucinations we find are often subtle, so we advise caution when using models to synthesize academic papers.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 17 of its 26 references that are in the map. First posted on arXiv 16 Oct 2023.

Builds on

17 of 26 references.

References · 17

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  3. 2017Deep reinforcement learning from human preferencescited by 3,912
  4. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  5. 2022Survey of Hallucination in Natural Language Generationcited by 2,075
  6. 2022Language Models (Mostly) Know What They Knowcited by 1,856
  7. 2016Concrete Problems in AI Safetycited by 1,642
  8. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 1,017
  9. 2020On Faithfulness and Factuality in Abstractive Summarizationcited by 900
  10. 2021Retrieval Augmentation Reduces Hallucination in Conversationcited by 692
  11. 2019Evaluating the Factual Consistency of Abstractive Text Summarizationcited by 584
  12. 2022Self-critiquing models for assisting human evaluatorscited by 380
5 more
  1. 2023Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedbackcited by 353
  2. 2023Chain-of-Verification Reduces Hallucination in Large Language Modelscited by 324
  3. 2021Hallucinated but Factual! Inspecting the Factuality of Hallucinations in Abstractive Summarizationcited by 129
  4. 2021The Factual Inconsistency Problem in Abstractive Text Summarization: A Surveycited by 79
  5. 2023Zero-shot Faithful Factual Error Correctioncited by 26

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 1.

4 papers.

  1. 2025FaithLens: Detecting and Explaining Faithfulness Hallucinationcited by 3
  2. 2025Confabulations from ACL Publications (CAP): A Dataset for Scientific Hallucination Detectioncited by 2
  3. 2025OpenFActScore: Open-Source Atomic Evaluation of Factuality in Text Generationcited by 1
  4. 2026Sharding Prevents LLM Oversight Failures and Adversarial Exploitationcited by 0