SPINE: Measuring LLM Sycophancy under Sustained Multi-Turn Pressure

arXiv:2609.09090 · cs.CL, cs.AI · Submitted 2026-09-08 · Read on arXiv

cs.CL, cs.AI

Submitted: 2026-09-08

Updated: 2026-10-13

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Sycophancy — 4× in this paper · explained in 4 episodes
• Trace — 2× in this paper · explained in 2 episodes
• Ablation — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Reasoning Traces — 1× in this paper · explained in 3 episodes

Sources

• A General Language Assistant as a Laboratory for Alignment
• DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
• SycEval: Evaluating LLM Sycophancy
• MONICA: Real-Time Monitoring and Calibration of Chain-of-Thought Sycophancy in Large Reasoning Models
• Training language models to be warm and empathetic makes them less reliable and more sycophantic
• Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experiment
• Consistency of Large Reasoning Models Under Multi-Turn Attacks
• The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning
• ICON: Intent-Context Coupling for Efficient Multi-Turn Jailbreak Attack
• TRUTH DECAY: Quantifying Multi-Turn Sycophancy in Language Models
• StereoSet: Measuring stereotypical bias in pretrained language models
• X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents
• LLMs know their vulnerabilities: Uncover Safety Gaps through Natural Distribution Shifts
• Great, Now Write an Article About That: The Crescendo Multi-Turn LLM Jailbreak Attack
• Towards Understanding Sycophancy in Language Models
• Olmo 3
• Sycophancy Is Not One Thing: Causal Separation of Sycophantic Behaviors in LLMs
• Persona Features Control Emergent Misalignment
• Foot-In-The-Door: A Multi-turn Jailbreak for LLMs
• CREPE: Open-Domain Question Answering with False Presuppositions

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering