# Sycophancy under Pressure: Evaluating and Mitigating Sycophantic Bias via Adversarial Dialogues in Scientific QA

arXiv: https://arxiv.org/abs/2508.13743 · HTML: https://mlsanity.com/paper/2508.13743/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 28 references that are in the map.
- First posted on arXiv 19 Aug 2025.

## Builds on (most cited first, 23 of 23)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [Gemma 3 Technical Report](https://mlsanity.com/paper/2503.19786/) (2025)
- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [Crowdsourcing Multiple Choice Science Questions](https://mlsanity.com/paper/1707.06209/) (2017)
- [ELEPHANT: Measuring and understanding social sycophancy in LLMs](https://mlsanity.com/paper/2505.13995/) (2025)
- [Measuring Sycophancy of Language Models in Multi-turn Dialogues](https://mlsanity.com/paper/2505.23840/) (2025)
- [Sycophancy in Large Language Models: Causes and Mitigations](https://mlsanity.com/paper/2411.15287/) (2024)
- [Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experiment](https://mlsanity.com/paper/2311.08596/) (2023)
- [From Yes-Men to Truth-Tellers: Addressing Sycophancy in Large Language Models with Pinpoint Tuning](https://mlsanity.com/paper/2409.01658/) (2024)
- [TRUTH DECAY: Quantifying Multi-Turn Sycophancy in Language Models](https://mlsanity.com/paper/2503.11656/) (2025)
- [Linear Probe Penalties Reduce LLM Sycophancy](https://mlsanity.com/paper/2412.00967/) (2024)
- [Language Models in Dialogue: Conversational Maxims for Human-AI Interactions](https://mlsanity.com/paper/2403.15115/) (2024)

## Built on by (most cited first, 18 of 18)

- [Sycophancy Is Not One Thing: Causal Separation of Sycophantic Behaviors in LLMs](https://mlsanity.com/paper/2509.21305/) (2025)
- [When Identity Skews Debate: Anonymization for Bias-Reduced Multi-Agent Reasoning](https://mlsanity.com/paper/2510.07517/) (2025)
- [What Counts as AI Sycophancy? A Taxonomy and Expert Survey of a Fragmented Construct](https://mlsanity.com/paper/2605.21778/) (2026)
- [MONICA: Real-Time Monitoring and Calibration of Chain-of-Thought Sycophancy in Large Reasoning Models](https://mlsanity.com/paper/2511.06419/) (2025)
- [Good Arguments Against the People Pleasers: How Reasoning Mitigates (Yet Masks) LLM Sycophancy](https://mlsanity.com/paper/2603.16643/) (2026)
- [SWAY: A Counterfactual Computational Linguistic Approach to Measuring and Mitigating Sycophancy](https://mlsanity.com/paper/2604.02423/) (2026)
- [When Correct Beliefs Collapse: Epistemic Resilience of LLMs under Clinical Pressure](https://mlsanity.com/paper/2605.23932/) (2026)
- [Recalling Too Well: Sycophancy Evaluation and Mitigation in Memory-Augmented Models](https://mlsanity.com/paper/2606.10949/) (2026)
- [Learning When to Trust via Selective Context Preference Optimization](https://mlsanity.com/paper/2608.06377/) (2026)
- [Evaluating Evidence Grounding Under User Pressure in Instruction-Tuned Language Models](https://mlsanity.com/paper/2603.20162/) (2026)
- [Who Flips? Self- and Cross-Model Counterarguments Reveal Answer Instability in LLMs](https://mlsanity.com/paper/2606.16011/) (2026)
- [Feedback Indices to Evaluate LLM Responses to Rebuttals for Multiple Choice Type Questions](https://mlsanity.com/paper/2601.03285/) (2026)
- [Agentic AI, Retrieval-Augmented Generation, and the Institutional Turn: Legal Architectures and Financial Governance in the Age of Distributional AGI](https://mlsanity.com/paper/2603.13244/) (2026)
- [Do Large Language Models Get Caught in Hofstadter-Mobius Loops?](https://mlsanity.com/paper/2603.13378/) (2026)
- [TD-DPO: Difference-Aware Preference Optimization for Mitigating Sycophancy in Clinical Autism Intervention Dialogue](https://mlsanity.com/paper/2607.18304/) (2026)
- [Mitigating LLM sycophancy with RL-based fine-tuning: Bayesian Truth Serum approach](https://mlsanity.com/paper/2608.25267/) (2026)
- [Unlearning Deceptive Behaviors in LLMs with Contrastive Forget Sets](https://mlsanity.com/paper/2609.38909/) (2026)
- [Beyond the Sycophancy Score: How Task, Model, and Pressure Shape LLM Yielding](https://mlsanity.com/paper/2610.08840/) (2026)

## Abstract (arXiv)

Large language models (LLMs), while increasingly used in domains requiring factual rigor, often display a troubling behavior: sycophancy, the tendency to align with user beliefs regardless of correctness. This tendency is reinforced by preference-based alignment techniques that optimize for user satisfaction but can undermine truthfulness. While relatively benign in casual dialogue, sycophancy poses serious risks in high-stakes settings such as scientific question answering (QA), where model outputs may shape collaborative reasoning, decision-making, and knowledge formation. Despite its importance, this phenomenon remains underexamined in factual QA contexts. We address this gap by introducing a unified evaluation framework to quantify the impact of sycophantic context on model behavior in scientific QA, measuring how much user-imposed social pressure distorts model outputs. The framework incorporates adversarial prompting setups and targeted metrics, such as misleading resistance and sycophancy resistance, that capture a model's ability to maintain factual consistency under misleading cues. Systematic evaluations across open-source and proprietary models reveal pervasive sycophantic tendencies, driven more by alignment strategy than by model size. To mitigate this issue, we propose Pressure-Tune, a lightweight post-training method that fine-tunes models on synthetic adversarial dialogues paired with chain-of-thought rationales. These rationales reject user misinformation while reinforcing factual commitments. Experiments on challenging scientific QA benchmarks show that Pressure-Tune significantly enhances sycophancy resistance without compromising accuracy or responsiveness to valid feedback, offering a practical pathway toward more truthful and principled model behavior.