mlsanityData to 14 Oct 2026

An Assessment of Model-On-Model Deception

arXiv 2405.12999 ↗First posted 10 May 2024cs.CL cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

The trustworthiness of highly capable language models is put at risk when they are able to produce deceptive outputs. Moreover, when models are vulnerable to deception it undermines reliability. In this paper, we introduce a method to investigate complex, model-on-model deceptive scenarios. We create a dataset of over 10,000 misleading explanations by asking Llama-2 7B, 13B, 70B, and GPT-3.5 to justify the wrong answer for questions in the MMLU. We find that, when models read these explanations, they are all significantly deceived. Worryingly, models of all capabilities are successful at misleading others, while more capable models are only slightly better at resisting deception. We recommend the development of techniques to detect and defend against deception.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 23 of its 31 references that are in the map. First posted on arXiv 10 May 2024.

Builds on

23 of 31 references.

References · 23

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  3. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  4. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  5. 2017Deep reinforcement learning from human preferencescited by 3,912
  6. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,211
  7. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,975
  8. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,336
  9. 2023Inference-Time Intervention: Eliciting Truthful Answers from a Language Modelcited by 1,302
  10. 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 836
  11. 2022Discovering Latent Knowledge in Language Models Without Supervisioncited by 767
  12. 2023The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasetscited by 712
11 more
  1. 2023Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Modelscited by 698
  2. 2023The Internal State of an LLM Knows When It's Lyingcited by 659
  3. 2024Hallucination is Inevitable: An Innate Limitation of Large Language Modelscited by 332
  4. 2023Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards and Ethical Behavior in the MACHIAVELLI Benchmarkcited by 182
  5. 2023Large Language Models can Strategically Deceive their Users when Put Under Pressurecited by 146
  6. 2023Deception Abilities Emerged in Large Language Modelscited by 101
  7. 2023Still No Lie Detector for Language Models: Probing Empirical and Conceptual Roadblockscited by 78
  8. 2023Hoodwinked: Deception and Cooperation in a Text-Based Game for Language Modelscited by 47
  9. 2023Honesty Is the Best Policy: Defining and Mitigating AI Deceptioncited by 40
  10. 2023Comparing Traditional and LLM-based Search for Consumer Choice: A Randomized Experimentcited by 36
  11. 2024Large Language Models vs. Search Engines: Evaluating User Preferences Across Varied Information Retrieval Scenarioscited by 14

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 1; 2026: 2.

5 papers.

  1. 2025Compromising Honesty and Harmlessness in Language Models via Deception Attackscited by 17
  2. 2024Lies, Damned Lies, and Distributional Language Statistics: Persuasion and Deception with Large Language Modelscited by 15
  3. 2024Moral Persuasion in Large Language Models: Evaluating Susceptibility and Ethical Alignmentcited by 10
  4. 2026Learning to Persuade Exposes How Easily LLMs Abandon Correct Beliefscited by 2
  5. 2026Probing the Limits of the Lie Detector Approach to LLM Deceptioncited by 0