mlsanityData to 14 Oct 2026

First Hallucination Tokens Are Different from Conditional Ones

arXiv 2507.20836 ↗First posted 28 Jul 2025cs.LG cs.AIcited by 4

Not on the map. Why?

Abstract (arXiv)

Large Language Models (LLMs) hallucinate, and detecting these cases is key to ensuring trust. While many approaches address hallucination detection at the response or span level, recent work explores token-level detection, enabling more fine-grained intervention. However, the distribution of hallucination signal across sequences of hallucinated tokens remains unexplored. We leverage token-level annotations from the RAGTruth corpus and find that the first hallucinated token is far more detectable than later ones. This structural property holds across models, suggesting that first hallucination tokens play a key role in token-level hallucination detection. Our code is available at https://github.com/jakobsnl/RAGTruth_Xtended.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 16 of its 23 references that are in the map. First posted on arXiv 28 Jul 2025.

Builds on

16 of 23 references.

References · 16

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2023A Survey of Hallucination in Large Foundation Modelscited by 368
  3. 2024INSIDE: LLMs' Internal States Retain the Power of Hallucination Detectioncited by 323
  4. 2023RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Modelscited by 283
  5. 2023Challenges and Applications of Large Language Modelscited by 254
  6. 2024Fine-grained Hallucination Detection and Editing for Language Modelscited by 149
  7. 2021A Token-level Reference-free Hallucination Detection Benchmark for Free-form Text Generationcited by 138
  8. 2023Do Language Models Know When They're Hallucinating References?cited by 116
  9. 2024The Dawn After the Dark: An Empirical Study on Factuality Hallucination in Large Language Modelscited by 88
  10. 2023On Early Detection of Hallucinations in Factual Question Answeringcited by 56
  11. 2024On Large Language Models' Hallucination with Regard to Known Factscited by 43
  12. 2024Detecting Hallucinations in Large Language Model Generation: A Token Probability Approachcited by 42
4 more
  1. 2024Scaling Up Membership Inference: When and How Attacks Succeed on Large Language Modelscited by 24
  2. 2024A Probabilistic Framework for LLM Hallucination Detection via Belief Tree Propagationcited by 21
  3. 2025Token-Level Density-Based Uncertainty Quantification Methods for Eliciting Truthfulness of Large Language Modelscited by 18
  4. 2024Detecting Hallucination and Coverage Errors in Retrieval Augmented Generation for Controversial Topicscited by 9

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026Hallucination as Commitment Failure: Larger LLMs Misfire Despite Knowing the Answercited by 2
  2. 2026How Confident Is the First Token? An Uncertainty-Calibrated Prompt Optimization Framework for Large Language Model Classification and Understandingcited by 1
  3. 2026Quickest Detection of Hallucination Onset: Delay Bounds and Learned CUSUM Statisticscited by 1
  4. 2026The Commit-Abstain Circuit: Why Language Models Hallucinate Instead of Abstainingcited by 0