# Sycophancy Is Not One Thing: Causal Separation of Sycophantic Behaviors in LLMs

arXiv: https://arxiv.org/abs/2509.21305 · HTML: https://mlsanity.com/paper/2509.21305/

## Key facts

- Cited by 37 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 45 references that are in the map.
- First posted on arXiv 25 Sep 2025.

## Builds on (most cited first, 26 of 26)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets](https://mlsanity.com/paper/2310.06824/) (2023)
- [Persona Vectors: Monitoring and Controlling Character Traits in Language Models](https://mlsanity.com/paper/2507.21509/) (2025)
- [AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders](https://mlsanity.com/paper/2501.17148/) (2025)
- [ELEPHANT: Measuring and understanding social sycophancy in LLMs](https://mlsanity.com/paper/2505.13995/) (2025)
- [Future Lens: Anticipating Subsequent Tokens from a Single Hidden State](https://mlsanity.com/paper/2311.04897/) (2023)
- [Measuring Sycophancy of Language Models in Multi-turn Dialogues](https://mlsanity.com/paper/2505.23840/) (2025)
- [LLMs Encode Harmfulness and Refusal Separately](https://mlsanity.com/paper/2507.11878/) (2025)
- [When Truth Is Overridden: Uncovering the Internal Origins of Sycophancy in Large Language Models](https://mlsanity.com/paper/2508.02087/) (2025)
- [Be Friendly, Not Friends: How LLM Sycophancy Shapes User Trust](https://mlsanity.com/paper/2502.10844/) (2025)
- [Linear Probe Penalties Reduce LLM Sycophancy](https://mlsanity.com/paper/2412.00967/) (2024)
- [Interaction Context Often Increases Sycophancy in LLMs](https://mlsanity.com/paper/2509.12517/) (2025)
- [MCPXKIT: The Unified Toolkit for Analyzing Model Context Protocol Security](https://mlsanity.com/paper/2508.12538/) (2025)
- [Benchmarking Deception Probes via Black-to-White Performance Boosts](https://mlsanity.com/paper/2507.12691/) (2025)
- [Sycophancy under Pressure: Evaluating and Mitigating Sycophantic Bias via Adversarial Dialogues in Scientific QA](https://mlsanity.com/paper/2508.13743/) (2025)
- [Flattering to Deceive: The Impact of Sycophantic Behavior on User Trust in Large Language Model](https://mlsanity.com/paper/2412.02802/) (2024)
- [Technological folie `a deux: Feedback Loops Between AI Chatbots and Mental Illness](https://mlsanity.com/paper/2507.19218/) (2025)
- ["Check My Work?": Measuring Sycophancy in a Simulated Educational Context](https://mlsanity.com/paper/2506.10297/) (2025)
- [The Geometry of Harmfulness in LLMs through Subconcept Probing](https://mlsanity.com/paper/2507.21141/) (2025)
- [How Do LLMs Persuade? Linear Probes Can Uncover Persuasion Dynamics in Multi-Turn Conversations](https://mlsanity.com/paper/2508.05625/) (2025)
- [Can You Trust an LLM with Your Life-Changing Decision? An Investigation into AI High-Stakes Responses](https://mlsanity.com/paper/2507.21132/) (2025)

## Built on by (most cited first, 30 of 37)

- [Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior](https://mlsanity.com/paper/2605.05115/) (2026)
- [What Counts as AI Sycophancy? A Taxonomy and Expert Survey of a Fragmented Construct](https://mlsanity.com/paper/2605.21778/) (2026)
- [LLMs Know They're Wrong and Agree Anyway: The Shared Sycophancy-Lying Circuit](https://mlsanity.com/paper/2604.19117/) (2026)
- [Good Arguments Against the People Pleasers: How Reasoning Mitigates (Yet Masks) LLM Sycophancy](https://mlsanity.com/paper/2603.16643/) (2026)
- [Verbalizing LLMs' assumptions to explain and control sycophancy](https://mlsanity.com/paper/2604.03058/) (2026)
- [Most LLM Conformity Needs No Speaker: Measuring the Speaker-Free Floor in Peer-Pressure Benchmarks](https://mlsanity.com/paper/2607.05545/) (2026)
- [Pressure, What Pressure? Sycophancy Disentanglement in Language Models via Reward Decomposition](https://mlsanity.com/paper/2604.05279/) (2026)
- [Playing Devil's Advocate: Off-the-Shelf Persona Vectors Rival Targeted Steering for Sycophancy](https://mlsanity.com/paper/2605.21006/) (2026)
- [Rhetorical Questions in LLM Representations: A Linear Probing Study](https://mlsanity.com/paper/2604.14128/) (2026)
- [Knowing but Not Correcting: Routine Task Requests Suppress Factual Correction in LLMs](https://mlsanity.com/paper/2605.05957/) (2026)
- [Not Just RLHF: Why Alignment Alone Won't Fix Multi-Agent Sycophancy](https://mlsanity.com/paper/2605.12991/) (2026)
- [Why LLMs Give In: Conversational Factors and Reasoning Behind Medical Sycophancy](https://mlsanity.com/paper/2608.01017/) (2026)
- [SPINE: Measuring LLM Sycophancy under Sustained Multi-Turn Pressure](https://mlsanity.com/paper/2609.09090/) (2026)
- [Reflections and New Directions for Human-Centered Large Language Models](https://mlsanity.com/paper/2605.06901/) (2026)
- [Sycophantic Praise: Evaluating Excessive Praise in Language Models](https://mlsanity.com/paper/2606.07441/) (2026)
- [Routing Subspaces: Auditing Evaluation-to-Deployment Mismatch in Fine-Tuned Language Models](https://mlsanity.com/paper/2607.20436/) (2026)
- [Sycophancy Suppression Can Impair Rational Updating: Anti-Sycophancy Should Preserve the Ability to Update](https://mlsanity.com/paper/2608.26511/) (2026)
- [Sycophantic Agreement Transfers with Neutral Data via Contrastive Preference Optimization](https://mlsanity.com/paper/2608.31079/) (2026)
- [Authority Bias in Language Models: Source Deference and User Agreement Are Not Interchangeable](https://mlsanity.com/paper/2609.37616/) (2026)
- [Making LLMs Reliable When It Matters Most: A Five-Layer Architecture for High-Stakes Decisions](https://mlsanity.com/paper/2511.07669/) (2025)
- [Who Flips? Self- and Cross-Model Counterarguments Reveal Answer Instability in LLMs](https://mlsanity.com/paper/2606.16011/) (2026)
- [Resist and Update: Counterfactual Report Coordinates for Incentive-Compatible LLMs](https://mlsanity.com/paper/2607.12985/) (2026)
- [Rewriting or Reweighting? A Geometric Account in Language Models](https://mlsanity.com/paper/2608.01835/) (2026)
- [One Axis, No Brake: Self-Knowledge Limits the Filtering of Harmful Peer Conformity in LLMs](https://mlsanity.com/paper/2609.18998/) (2026)
- [Aligning to Illusions: Choice Blindness in Human and AI Feedback](https://mlsanity.com/paper/2603.08412/) (2026)
- [Auditing Framing-Sensitive Behavioral Instability in Large Language Models for Mental Health Interactions](https://mlsanity.com/paper/2606.26982/) (2026)
- [Relational Positioning as a Measurable Risk Object: History-Carried Lock-in and Self-Confabulation in Multi-Turn Human-AI Dialogue](https://mlsanity.com/paper/2607.11437/) (2026)
- [PCA-guided Activation Scaling for Monotonic Bidirectional Control over LLM Sycophancy](https://mlsanity.com/paper/2608.16650/) (2026)
- [PCFBench: A Diagnostic Benchmark for Product Carbon Footprint Estimation](https://mlsanity.com/paper/2608.27716/) (2026)
- [Emotional Labor Strategy Preferences in LLM Personas](https://mlsanity.com/paper/2609.00310/) (2026)

## Abstract (arXiv)

Large language models (LLMs) often exhibit sycophantic behaviors -- such as excessive agreement with or flattery of the user -- but it is unclear whether these behaviors arise from a single mechanism or multiple distinct processes. We decompose sycophancy into sycophantic agreement and sycophantic praise, contrasting both with genuine agreement. Using difference-in-means directions, activation additions, and subspace geometry across multiple models and datasets, we show that: (1) the three behaviors are encoded along distinct linear directions in latent space; (2) each behavior can be independently amplified or suppressed without affecting the others; and (3) their representational structure is consistent across model families and scales. These results suggest that sycophantic behaviors correspond to distinct, independently steerable representations.