# Beyond Social Pressure: Benchmarking Epistemic Attack in Large Language Models

arXiv: https://arxiv.org/abs/2604.07749 · HTML: https://mlsanity.com/paper/2604.07749/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 24 references that are in the map.
- First posted on arXiv 9 Apr 2026.

## Builds on (most cited first, 16 of 16)

- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [Representation Engineering: A Top-Down Approach to AI Transparency](https://mlsanity.com/paper/2310.01405/) (2023)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [In-Context Impersonation Reveals Large Language Models' Strengths and Biases](https://mlsanity.com/paper/2305.14930/) (2023)
- [Measuring Sycophancy of Language Models in Multi-turn Dialogues](https://mlsanity.com/paper/2505.23840/) (2025)
- [Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning](https://mlsanity.com/paper/2512.20848/) (2025)
- [Mind the Value-Action Gap: Do LLMs Act in Alignment with Their Values?](https://mlsanity.com/paper/2501.15463/) (2025)
- [SocREval: Large Language Models with the Socratic Method for Reference-Free Reasoning Evaluation](https://mlsanity.com/paper/2310.00074/) (2023)
- [MindGames: Targeting Theory of Mind in Large Language Models with Dynamic Epistemic Modal Logic](https://mlsanity.com/paper/2305.03353/) (2023)
- [Sycophancy in Vision-Language Models: A Systematic Analysis and an Inference-Time Mitigation Framework](https://mlsanity.com/paper/2408.11261/) (2024)
- [Epistemic Injustice in Generative AI](https://mlsanity.com/paper/2408.11441/) (2024)
- [Are BabyLMs Deaf to Gricean Maxims? A Pragmatic Evaluation of Sample-efficient Language Models](https://mlsanity.com/paper/2510.04764/) (2025)

## Built on by (most cited first, 1 of 1)

- [MEDLEY-BENCH: Benchmarking Behavioural Metacognition and Belief Revision Under Social Pressure in Large Language Models](https://mlsanity.com/paper/2604.16009/) (2026)

## Abstract (arXiv)

Large language models (LLMs) can shift their answers under pressure in ways that reflect accommodation rather than reasoning. Prior work on sycophancy has focused mainly on disagreement, flattery, and preference alignment, leaving a broader set of epistemic failures less explored. We introduce PPT-Bench, a diagnostic benchmark for evaluating textitepistemic attack, where prompts challenge the legitimacy of knowledge, values, or identity rather than simply opposing a previous answer. PPT-Bench is organized around the Philosophical Pressure Taxonomy (PPT), which defines four types of philosophical pressure: Epistemic Destabilization, Value Nullification, Authority Inversion, and Identity Dissolution. Each item is tested at three layers: a baseline prompt (L0), a single-turn pressure condition (L1), and a multi-turn Socratic escalation (L2). This allows us to measure epistemic inconsistency between L0 and L1, and conversational capitulation in L2. Across five models, these pressure types produce statistically separable inconsistency patterns, suggesting that epistemic attack exposes weaknesses not captured by standard social-pressure benchmarks. Mitigation results are strongly type- and model-dependent: prompt-level anchoring and persona-stability prompts perform best in API settings, while Leading Query Contrastive Decoding is the most reliable intervention for open models.