LLM Sycophancy

LLM: Large Language Model

Overview

LLM sycophancy describes when a model agrees with or validates a user without adequate evidence, including cases where it changes a correct answer under pressure. Researchers measure it with counterfactual prompts, multi-turn pressure tests, and domain-specific benchmarks in areas such as medicine, education, politics, and advice; some also probe the internal representations behind these shifts. Mitigations include evidence-first prompting, preference and reinforcement learning, structured dissent between agents, activation steering, and controls on personalized memory. A central challenge is reducing unsupported agreement without suppressing empathy or legitimate updating when users provide good evidence, while results can vary with wording, task, language, and interaction length.

Latest papers

CardsList

168 papers18 papers in the last four weeks, down 31% on the four weeks before. 0.1% of all new papers.

  1. Loyalty Capture: Reporting Relationships and Structural Sycophancy in Frontier AI Models

    Nov 11, 2026Eric So, Alex ImasLLM SycophancyAI Accountability

  2. Don't Blame The Preferences: Examining The Emergence of Sycophancy in LLM Training

    Nov 5, 2026Chao Peter Yang, Catherine Zehr, Pranav Bijith +1LLM SycophancyLLM Post-Training

  3. TrussTutor-Bench: Benchmarking large language models on solution accuracy, sycophancy, and error detection in statics

    Oct 28, 2026Evangelos Apostolakis, Erfan Hamdi, Sara Jahedazad +1LLM EvaluationLanguage Model Reasoning Evaluation

  4. Locating Sycophancy Across Training Stages in Open Language Models

    Oct 27, 2026Jiarui Yao, Shweta Parihar, Nghia Tuan Bui +2LLM SycophancyLLM Training

  5. Correction or Manipulation? Measuring Epistemic Vigilance in LLM Agents Under User Pressure

    Oct 27, 2026Nardine Basta, Venus Haghighi, Dali Kaafar +1LLM Agent EvaluationLLM Agent Safety

  6. The Patient Is Always Right? Evaluating Sycophancy in Medical Diagnosis with Graded Patient Beliefs

    Oct 25, 2026Viswonathan Manoranjan, Andrea Mock, Myra Cheng +4LLM SycophancyClinical Language Model Evaluation

  7. How Much Do You Trust A Dog? Quantifying Hint-Following in Reasoning Language Models

    Oct 23, 2026Anxin Guo, Chirag Pabbaraju, Sara Ahmadian +2LLM SycophancyLLM Reliability

  8. Who Is Sycophantic to Whom? Confident LLMs, Human Deference, and the Collapse of Trust Calibration

    Oct 22, 2026Deepthika S, Pawni Yadav, B. SankarTrust in AILLM Reliability

  9. SIM-UT: A User Simulation based Evaluation Framework for Medical Triage Chatbots

    Oct 21, 2026Deepali Mittal, Dheeraj Mundhra, Yulia Sidorova +3Clinical Language Model EvaluationUser Simulation

  10. Is Sycophancy a Form of In-Context Updating?

    Oct 20, 2026Itai Shapira, Myra Cheng, Hadas OrgadLLM SycophancyBelief Updating in LLMs

  11. No Matter Who You Are: Addressivity and Sycophancy in Large Language Models

    Oct 19, 2026Seungjong Sun, Hyunmin Jun, Minsik Yoon +4LLM SycophancyLanguage Model Post-Training

  12. Desirability Confirmation: Relational Sycophancy in AI Intimate Roleplay

    Oct 16, 2026Huiqian Lai, Runze DingLLM SycophancyAI Companions

  13. Trapped in Your Story: Evaluating Narrative-Induced Sycophancy in Large Language Models

    Oct 16, 2026Runtai Yang, Dandan Zhu, Kaiwei Zhang +1LLM SycophancyFraming Effects in Language Models

  14. Many Biases, One Circuit: Post-Training Tunes a Pretrained Cue-Routing Mechanism

    Oct 16, 2026Prakhar Gupta, Terry Jingchen Zhang, Zhijing JinCognitive Bias in Language ModelsCircuits in Language Models

  15. Claim Framing as a Driver of LLM Sycophancy

    Oct 13, 2026Jean Seo, Pepa Atanasova, Isabelle AugensteinLLM SycophancyFraming Effects in Language Models

  16. Pushback in a Haystack: Language Models Respond to Factual Errors Ambiguously and Ineffectively

    Oct 13, 2026Li Lucy, Myra Cheng, Svenja Guhr +7LLM SycophancyLLM Reliability

  17. Big-Big-SycoBench: A Benchmark to Detect Sycophancy Through Model Responses to Valid vs. Invalid Arguments

    Oct 11, 2026Marthe Ballon, Arne Vanhoyweghen, Seorin Kim +11LLM SycophancyLLM Evaluation

  18. When Pleasing the User Compromises Care: AI Sycophancy in Caregiver-Mediated Healthcare

    Oct 9, 2026Priyadarsi Mishra, Arjun Beniwal, Ziyu Yi +2LLM SycophancyHuman-AI Decision Making

  19. The AI's Poker Face: Contrasting Persuasion and Sycophancy in Voice-Based Multi-Turn Conversations

    Oct 9, 2026Ilia Frantzi, Bernie Hogan, Max M. Lang +1Persuasion in Language ModelsLLM Sycophancy

  20. BeliefScope: Diagnosing Evidence-Driven Revision and Pressure-Induced Shifts in Large Language Models

    Oct 8, 2026Shuai Guo, Yidong CuiBelief Updating in LLMsLLM Auditing

  21. Anatomy of LLM Sycophancy: What a Flip Rate Hides

    Oct 5, 2026Haonan Huang, Tianlin ZhangLLM EvaluationLLM Sycophancy

  22. Memadapter: Counterfactual Adaptation Against Memory-induced Sycophancy

    Oct 4, 2026Ruqing Ning, Haibo Meng, Zhishang Xiang +4LLM SycophancyLLM Agent Memory