mlsanityData to 14 Oct 2026

Too Polite to Disagree: Understanding Sycophancy Propagation in Multi-Agent Systems

arXiv 2604.02668 ↗First posted 3 Apr 2026cs.CL cs.AI cs.MAcited by 6

On the map: LLM agents and tool use → Social simulation and multi-agent debate

Abstract (arXiv)

Large language models (LLMs) often exhibit sycophancy: agreement with user stance even when it conflicts with the model's opinion. While prior work has mostly studied this in single-agent settings, it remains underexplored in collaborative multi-agent systems. We ask whether awareness of other agents' sycophancy levels influences discussion outcomes. To investigate this, we run controlled experiments with six open-source LLMs, providing agents with peer sycophancy rankings that estimate each peer's tendency toward sycophancy. These rankings are based on scores calculated using various static (pre-discussion) and dynamic (online) strategies. We find that providing sycophancy priors reduces the influence of sycophancy-prone peers, mitigates error-cascades, and improves final discussion accuracy by an absolute 10.5%. Thus, this is a lightweight and efficient way to reduce model sycophancy during discussions and subsequently improve downstream accuracy.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use. Builds on 15 of its 28 references that are in the map. First posted on arXiv 3 Apr 2026.

Builds on

15 of 28 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025When Identity Skews Debate: Anonymization for Bias-Reduced Multi-Agent Reasoning8 Oct
  2. 2025Peacemaker or Troublemaker: How Sycophancy Shapes Multi-Agent Debate27 Sep
  3. 2025Talk Isn't Always Cheap: Understanding Failure Modes in Multi-Agent Debate5 Sep

Other references · 12

  1. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  2. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  3. 2023Towards Understanding Sycophancy in Language Modelscited by 1,040
  4. 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 836
  5. 2023ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMscited by 309
  6. 2025SycEval: Evaluating LLM Sycophancycited by 108
  7. 2025Measuring Sycophancy of Language Models in Multi-turn Dialoguescited by 104
  8. 2024Sycophancy in Large Language Models: Causes and Mitigationscited by 101
  9. 2023Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experimentcited by 77
  10. 2025Do as We Do, Not as You Think: the Conformity of Large Language Modelscited by 53
  11. 2024Number Cookbook: Number Understanding of Language Models and How to Improve Itcited by 42
  12. 2025Sycophancy Mitigation Through Reinforcement Learning with Uncertainty-Aware Adaptive Reasoning Trajectoriescited by 11

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 6.

6 papers.

  1. 2026Flag Game: A Toy Model for Mechanistic Swarm Interpretabilitycited by 2
  2. 2026PredAct-Bench: Benchmarking Tool-Augmented Dialogue under Controlled Tool Noisecited by 0
  3. 2026What Does Multi-Agent LLM Debate Actually Change? A Layered Analysis of Disagreement and Answer Qualitycited by 0
  4. 2026"You're Right, Let Me Fix It": How LLM Agents Damage Correct Work When Falsely Accusedcited by 0
  5. 2026Measuring Collapse and Correction in Homogeneous-Panel LLM Debatecited by 0
  6. 2026Review Rarely Beats the Stronger Model: Separating Verification from Capability Transfer in LLM Pipelinescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Agentic Scaffolding Amplifies Sycophantic Behavior in Large Language Models
  2. 2025When Truth Is Overridden: Uncovering the Internal Origins of Sycophancy in Large Language Models
  3. 2026Not Just RLHF: Why Alignment Alone Won't Fix Multi-Agent Sycophancy
  4. 2026SPINE: Measuring LLM Sycophancy under Sustained Multi-Turn Pressure
  5. 2025An Empirical Study of Group Conformity in Multi-Agent Systems
  6. 2025Sycophancy under Pressure: Evaluating and Mitigating Sycophantic Bias via Adversarial Dialogues in Scientific QA
  7. 2026Collective Opinion Dynamics in Structured LLM Populations
  8. 2026When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Models