Sycophantic Anchors: Localizing and Quantifying User Agreement in Reasoning Models
On the map: LLM reasoning → Prompting and chain-of-thought reasoning
Abstract (arXiv)
Reasoning models frequently agree with incorrect user suggestions -- a behavior known as sycophancy. However, it is unclear where in the reasoning trace this agreement originates and how strong the commitment is. We introduce sycophantic anchors -- sentences identified via counterfactual analysis that commit models to user agreement. Across four reasoning models spanning three architecture families (Llama, Qwen, Falcon-hybrid) and 1.5B--8B parameters, we analyze over 200,000 counterfactual rollouts and show that linear probes reliably detect sycophantic anchors (74--85% balanced accuracy), outperforming text-only baselines at high commitment levels -- confirming they capture internal states beyond surface vocabulary. Regressors further predict commitment strength from activations (R2 up to 0.74). We observe a consistent asymmetry: sycophancy leaves a stronger mechanistic footprint than correct reasoning. We also find that sycophancy builds gradually during generation rather than being determined by the prompt. These findings enable sentence-level detection and quantification of model misalignment mid-inference.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 15 of its 21 references that are in the map. First posted on arXiv 29 Jan 2026.
Builds on
15 of 21 references.
Closest ancestors · 2
Same direction, newest first
- 2025Thought Anchors: Which LLM Reasoning Steps Matter?23 Jun
- 2023Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting7 May
Other references · 13
- 2017Attention Is All You Needcited by 59,932
- 2024The Llama 3 Herd of Modelscited by 14,192
- 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
- 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,710
- 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Dualitycited by 1,283
- 2023Inference-Time Intervention: Eliciting Truthful Answers from a Language Modelcited by 1,259
- 2023Towards Understanding Sycophancy in Language Modelscited by 988
- 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 805
- 2022Discovering Latent Knowledge in Language Models Without Supervisioncited by 733
- 2024Personalized Large Language Modelscited by 44
- 2026Falcon-H1R: Pushing the Reasoning Frontiers with a Hybrid Model for Efficient Test-Time Scalingcited by 12
- 2025MONICA: Real-Time Monitoring and Calibration of Chain-of-Thought Sycophancy in Large Reasoning Modelscited by 10
1 more
Built on by
Citing papers per year: 2026: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2025Internal Reasoning vs. External Control: A Thermodynamic Analysis of Sycophancy in Large Language Models
- 2025Measuring Sycophancy of Language Models in Multi-turn Dialogues
- 2026Agentic Scaffolding Amplifies Sycophantic Behavior in Large Language Models
- 2026When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Models
- 2024Sycophancy in Large Language Models: Causes and Mitigations
- 2026LLMs Know They're Wrong and Agree Anyway: The Shared Sycophancy-Lying Circuit
- 2026Beyond the Sycophancy Score: How Task, Model, and Pressure Shape LLM Yielding
- 2026The Silicon Mirror: Dynamic Behavioral Gating for Anti-Sycophancy in LLM Agents