Suppressing Pressure, Amplifying Evidence: Self-Guided Attention Steering to Mitigate Sycophancy and Stubbornness
cs.AI, cs.CL
Submitted: 2026-10-03
Updated: 2026-10-03
Code: https://github.com/03Grant/sycophancy-and-stubbornness
Terminology
Sources
- Measuring Sycophancy of Language Models in Multi-turn Dialogues
- MONICA: Real-Time Monitoring and Calibration of Chain-of-Thought Sycophancy in Large Reasoning Models
- Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experiment
- Sycophancy Suppression Can Impair Rational Updating: Anti-Sycophancy Should Preserve the Ability to Update
- Inverse Scaling: When Bigger Isn't Better
- Gemma 4 Technical Report
- Simple synthetic data reduces sycophancy in large language models
- System 2 Attention (is something you might need too)
- Equitable System-Prompt Selection via Constrained Mixed-Strategy GroupDRO
- MIRA: A Bilingual Benchmark for Medical Information Response Audit
- Towards Reliable LLM Evaluation: Correcting the Winner's Curse in Adaptive Benchmarking
- Qwen2.5 Technical Report
- Model Tells Itself Where to Attend: Faithfulness Meets Automatic Attention Steering
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems