mlsanityData to 14 Oct 2026

TRUTH DECAY: Quantifying Multi-Turn Sycophancy in Language Models

arXiv 2503.11656 ↗First posted 4 Feb 2025cs.CLcited by 33

Not on the map. Why?

Abstract (arXiv)

Rapid improvements in large language models have unveiled a critical challenge in human-AI interaction: sycophancy. In this context, sycophancy refers to the tendency of models to excessively agree with or flatter users, often at the expense of factual accuracy. While previous studies have primarily analyzed this behavior in single-turn interactions, its persistence and evolution in multi-step conversations remain largely unexplored. We introduce TRUTH DECAY, a benchmark specifically designed to evaluate sycophancy in extended dialogues, where language models must navigate iterative user feedback, challenges, and persuasion. We prompt models to elicit four types of sycophantic biases. We then propose and test sycophancy reduction strategies, evaluating their effectiveness beyond single-step interactions.

Key facts. Cited by 33 papers in the map (data through 14 Oct 2026). Builds on 18 of its 25 references that are in the map. First posted on arXiv 4 Feb 2025.

Builds on

18 of 25 references.

References · 18

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  4. 2017Deep reinforcement learning from human preferencescited by 3,912
  5. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,975
  6. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,848
  7. 2023Towards Understanding Sycophancy in Language Modelscited by 1,040
  8. 2023Large Language Modelscited by 780
  9. 2023MINT: Evaluating LLMs in Multi-turn Interaction with Tools and Language Feedbackcited by 297
  10. 2024Sycophancy in Large Language Models: Causes and Mitigationscited by 101
  11. 2023Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experimentcited by 77
  12. 2024Anchoring Bias in Large Language Models: An Experimental Studycited by 37
6 more
  1. 2023DialogBench: Evaluating LLMs as Human-like Dialogue Systemscited by 31
  2. 2024Language Models in Dialogue: Conversational Maxims for Human-AI Interactionscited by 21
  3. 2022UX Research on Conversational Human-AI Interaction: A Literature Review of the ACM Digital Librarycited by 20
  4. 2024CBEval: A framework for evaluating and interpreting cognitive biases in LLMscited by 12
  5. 2024Chaos with Keywords: Exposing Large Language Models Sycophantic Hallucination to Misleading Keywords and Evaluating Defense Strategiescited by 10
  6. 2024Fool Me, Fool Me: User Attitudes Toward LLM Falsehoodscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 8; 2026: 25.

33 papers.

Sort:

By area:

  1. 2025Measuring Sycophancy of Language Models in Multi-turn Dialoguescited by 104
  2. 2025When Identity Skews Debate: Anonymization for Bias-Reduced Multi-Agent Reasoningcited by 21
  3. 2025Sycophancy under Pressure: Evaluating and Mitigating Sycophantic Bias via Adversarial Dialogues in Scientific QAcited by 18
  4. 2025Challenging the Evaluator: LLM Sycophancy Under User Rebuttalcited by 18
  5. 2026What Counts as AI Sycophancy? A Taxonomy and Expert Survey of a Fragmented Constructcited by 16
  6. 2025MONICA: Real-Time Monitoring and Calibration of Chain-of-Thought Sycophancy in Large Reasoning Modelscited by 12
  7. 2026MemSyco-Bench: Benchmarking Sycophancy in Agent Memorycited by 12
  8. 2026Good Arguments Against the People Pleasers: How Reasoning Mitigates (Yet Masks) LLM Sycophancycited by 8
  9. 2026When Correct Beliefs Collapse: Epistemic Resilience of LLMs under Clinical Pressurecited by 6
  10. 2026Attractor States Emerge in Multi-Turn LLM Conversationscited by 6
  11. 2026Personalization Increases Affective Alignment but Has Role-Dependent Effects on Epistemic Independence in LLMscited by 5
  12. 2026Most LLM Conformity Needs No Speaker: Measuring the Speaker-Free Floor in Peer-Pressure Benchmarkscited by 5
21 more
  1. 2025But what is your honest answer? Aiding LLM-judges with honest alternatives using steering vectorscited by 3
  2. 2025Cognitive models can reveal interpretable value trade-offs in language modelscited by 3
  3. 2026SPINE: Measuring LLM Sycophancy under Sustained Multi-Turn Pressurecited by 3
  4. 2026MedPRESS: A Multi-turn Benchmark for Patient-Pressure-Induced Medical Sycophancy in LLMscited by 2
  5. 2025Assertion-Conditioned Compliance: A Provenance-Aware Vulnerability in Multi-Turn Tool-Calling Agentscited by 1
  6. 2026Are LLMs Vulnerable to Preference-Undermining Attacks (PUA)? A Factorial Analysis Methodology for Diagnosing the Trade-off between Preference Alignment and Real-World Validitycited by 1
  7. 2026Stress Tests REVEAL Fragile Temporal and Visual Grounding in Video-Language Modelscited by 1
  8. 2026The Hidden Cost of Contextual Sycophancy: an AI Literacy Intervention in Human-AI Collaborationcited by 1
  9. 2026Machine Psychometrics: A Mathematical Psychology of Artificial Intelligencecited by 1
  10. 2026Who Flips? Self- and Cross-Model Counterarguments Reveal Answer Instability in LLMscited by 1
  11. 2026Token-Level Diagnosis of Sycophancy in LLMs with Attribution-Guided Steeringcited by 1
  12. 2026Responsible AI for General-Purpose Systems: Overview, Challenges, and A Path Forwardcited by 0
  13. 2026Squish and Release: Exposing Hidden Hallucinations by Making Them Surface as Safety Signalscited by 0
  14. 2026When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Modelscited by 0
  15. 2026ReCrit: Transition-Aware Reinforcement Learning for Scientific Critic Reasoningcited by 0
  16. 2026TD-DPO: Difference-Aware Preference Optimization for Mitigating Sycophancy in Clinical Autism Intervention Dialoguecited by 0
  17. 2026Measuring and Detecting Harmful AI Sycophancycited by 0
  18. 2026Caught in the Story: Narrative Captivity in Multi-turn LLMs Conversationcited by 0
  19. 2026How AI Models Manage Epistemic Authority: A Taxonomy and Comparative Analysis of Responses to User Disagreementcited by 0
  20. 2026Not All Answers Are Contextually Persuadable: Inference Dynamics in Large Language Models under Contextual Influencecited by 0
  21. 2026Memadapter: Counterfactual Adaptation Against Memory-induced Sycophancycited by 0