A Mechanistic View of Authority Hierarchy in LLM Sycophancy

• URL: http://arxiv.org/abs/2607.00415v1
• Date: Wed, 01 Jul 2026 04:16:59 GMT
• Title: A Mechanistic View of Authority Hierarchy in LLM Sycophancy
• Abstract summary: We show that authority-induced sycophancy is not a surface-level output bias but mechanistic knowledge erasure.<n>Our findings suggest that authority-induced sycophancy is not a surface-level output bias but mechanistic knowledge erasure.
• Score: 0.09085204695117637
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Authority bias poses a critical safety concern in language models: models systematically prioritize social cues from authority figures over factual consistency, swaying their answers based on source credibility rather than evidence. We mechanistically investigate this phenomenon using a controlled medical QA setting, where hints suggesting incorrect answers are attributed to personas of varying expertise. Across Llama-3.1-8B, Qwen3-8B, and Gemma-2-9B, we find that models respond in a graded manner proportional to perceived authority, a hierarchy that is never explicitly prompted but emerges from training. Logit lens analysis and linear/non-linear probing localize this effect to a critical late layer where correct answer representations are actively erased, an erasure that scales with authority level, resists mean vector intervention, and is only partially reversible through chain-of-thought reasoning. Our findings suggest that authority-induced sycophancy is not a surface-level output bias but mechanistic knowledge erasure, a precise, layer-localized overwriting of correct internal representations by high-status authority signals.

Related papers

• Token-Level Diagnosis of Sycophancy in LLMs with Attribution-Guided Steering [19.708436827458744]
  Sycophancy refers to the tendency for large language models to match user beliefs at the cost of factual correctness.<n>We introduce the Authority Share Index (ASI), which measures the degree to which a model's decision is driven by authority-related text.<n>We find that sycophantic responses consistently direct more attention toward authority tokens than resistant ones.
  arXiv  Detail & Related papers  (2026-07-31T00:05:36Z)
• Perception Without Engagement: Dissecting the Causal Discovery Deficit in LMMs [68.58207076756237]
  We introduce ProCauEval, a perturbation-based evaluation protocol that shifts from outcome assessment to mechanism diagnosis.<n>We find that models faithfully perceive video content yet systematically underexploit it during causal reasoning.<n>We propose Anti-Distillation Policy Optimization (ADPO), a reinforcement learning framework built on negative teacher alignment.
  arXiv  Detail & Related papers  (2026-05-10T08:48:58Z)
• Taming Actor-Observer Asymmetry in Agents via Dialectical Alignment [59.536125286960186]
  Multi-agent frameworks assigning specialized roles are increasingly adopted to enable self-reflection and mutual auditing.<n>We find it simultaneously induces a human-like cognitive bias known as Actor-Observer Asymmetry.<n>ReTAS guides agents to synthesize conflicting viewpoints into an objective consensus.
  arXiv  Detail & Related papers  (2026-04-21T15:05:58Z)
• METER: Evaluating Multi-Level Contextual Causal Reasoning in Large Language Models [61.33372454250959]
  Contextual causal reasoning is a critical yet challenging capability for Large Language Models.<n>Existing benchmarks fail to ensure context consistency or cover the full causal hierarchy.<n>We pioneer METER to systematically benchmark LLMs across all three levels of the causal ladder.
  arXiv  Detail & Related papers  (2026-04-13T14:07:11Z)
• Aligned Agents, Biased Swarm: Measuring Bias Amplification in Multi-Agent Systems [27.694287638709643]
  Real-world Multi-Agent Systems are too complex to analyze entirely.<n>We conduct a baseline empirical study investigating how basic MAS topologies and feedback loops influence prejudice.<n>By stripping away advanced complexity to study swarm dynamics, we establish a crucial baseline: structural complexity does not guarantee ethical robustness.
  arXiv  Detail & Related papers  (2026-04-10T05:06:38Z)
• Decoding the Critique Mechanism in Large Reasoning Models [50.821607345799386]
  Large Reasoning Models (LRMs) exhibit backtracking and self-verification mechanisms that enable them to revise intermediate steps and reach correct solutions.<n>We investigate how current LRMs recover from errors by inserting arithmetic mistakes in their intermediate reasoning steps.<n>We find that despite the error propagating through the chain-of-thought, the model still reaches the correct final answer.
  arXiv  Detail & Related papers  (2026-03-17T10:03:30Z)
• From Flat Logs to Causal Graphs: Hierarchical Failure Attribution for LLM-based Multi-Agent Systems [13.588125015284257]
  We propose CHIEF, a novel framework that transforms chaotic trajectories into a structured hierarchical causal graph.<n> Experiments on Who&When benchmark show that CHIEF outperforms eight strong and state-of-the-art baselines on both agent- and step-level accuracy.
  arXiv  Detail & Related papers  (2026-02-27T06:08:42Z)
• Agentic Uncertainty Quantification [76.94013626702183]
  We propose a unified Dual-Process Agentic UQ (AUQ) framework that transforms verbalized uncertainty into active, bi-directional control signals.<n>Our architecture comprises two complementary mechanisms: System 1 (Uncertainty-Aware Memory, UAM), which implicitly propagates verbalized confidence and semantic explanations to prevent blind decision-making; and System 2 (Uncertainty-Aware Reflection, UAR), which utilizes these explanations as rational cues to trigger targeted inference-time resolution only when necessary.
  arXiv  Detail & Related papers  (2026-01-22T07:16:26Z)
• When Truth Is Overridden: Uncovering the Internal Origins of Sycophancy in Large Language Models [11.001042171551566]
  We study how user opinions induce sycophancy across different model families.<n>First-person prompts consistently induce higher sycophancy rates than third-person framings.<n>These findings highlight that sycophancy is not a surface-level artifact but emerges from a structural override of learned knowledge in deeper layers.
  arXiv  Detail & Related papers  (2025-08-04T05:55:06Z)
• Sycophancy in Vision-Language Models: A Systematic Analysis and an Inference-Time Mitigation Framework [18.54098084470481]
  We analyze sycophancy across vision-language benchmarks and propose an inference-time mitigation framework.<n>Our framework effectively mitigates sycophancy across all evaluated models, while maintaining performance on neutral prompts.
  arXiv  Detail & Related papers  (2024-08-21T01:03:21Z)
• Understanding, Predicting and Better Resolving Q-Value Divergence in Offline-RL [86.0987896274354]
  We first identify a fundamental pattern, self-excitation, as the primary cause of Q-value estimation divergence in offline RL. We then propose a novel Self-Excite Eigenvalue Measure (SEEM) metric to measure the evolving property of Q-network at training. For the first time, our theory can reliably decide whether the training will diverge at an early stage.
  arXiv  Detail & Related papers  (2023-10-06T17:57:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.