Imagen de Portada

When Prohibitions Become Permissions: Auditing Negation Sensitivity in Language Models

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Elkins, Katherine, Chun, Jon
Formato:	Preprint
Publicado:	2026
Materias:	Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Syntactic Framing Fragility: An Audit of Robustness in LLM Ethical Decisions
por: Elkins, Katherine, et al.
Publicado: (2025)

The Paradox of Robustness: Decoupling Rule-Based Logic from Affective Noise in High-Stakes Decision-Making
por: Chun, Jon, et al.
Publicado: (2026)

Informed AI Regulation: Comparing the Ethical Frameworks of Leading LLM Chatbots Using an Ethics-Based Audit to Assess Moral Reasoning and Normative Values
por: Chun, Jon, et al.
Publicado: (2024)

AgenticSimLaw: A Juvenile Courtroom Multi-Agent Debate Simulation for Explainable High-Stakes Tabular Decision Making
por: Chun, Jon, et al.
Publicado: (2026)

The AI Fiction Paradox
por: Elkins, Katherine
Publicado: (2026)

Permissive Information-Flow Analysis for Large Language Models
por: Siddiqui, Shoaib Ahmed, et al.
Publicado: (2024)

Comparative Global AI Regulation: Policy Perspectives from the EU, China, and the US
por: Chun, Jon, et al.
Publicado: (2024)

When Prompt Optimization Becomes Jailbreaking: Adaptive Red-Teaming of Large Language Models
por: Shamsi, Zafir, et al.
Publicado: (2026)

Useful Memories Become Faulty When Continuously Updated by LLMs
por: Zhang, Dylan, et al.
Publicado: (2026)

Quantifying Gender Bias in Large Language Models: When ChatGPT Becomes a Hiring Manager
por: Gerszberg, Nina, et al.
Publicado: (2026)

When Emotion Becomes Trigger: Emotion-style dynamic Backdoor Attack Parasitising Large Language Models
por: Liu, Ziyu, et al.
Publicado: (2026)

Permissioned LLMs: Enforcing Access Control in Large Language Models
por: Jayaraman, Bargav, et al.
Publicado: (2025)

When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Models
por: Li, Jiechen, et al.
Publicado: (2026)

Generative Adversarial Reviews: When LLMs Become the Critic
por: Bougie, Nicolas, et al.
Publicado: (2024)

Permissive-Washing in the Open AI Supply Chain: A Large-Scale Audit of License Integrity
por: Jewitt, James, et al.
Publicado: (2026)

When Individually Calibrated Models Become Collectively Miscalibrated
por: Wang, Zhaohui
Publicado: (2026)

Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention
por: Rabanser, Stephan, et al.
Publicado: (2025)

PromptAudit: Auditing Prompt Sensitivity in LLM-Based Vulnerability Detection
por: Camarato, Steffen J., et al.
Publicado: (2026)

Permissible Knowledge Pooling
por: Dong, Huimin
Publicado: (2024)

Agentive Permissions in Multiagent Systems
por: Shi, Qi
Publicado: (2024)

Judicial Permission
por: Governatori, Guido, et al.
Publicado: (2025)

AuditLLM: A Tool for Auditing Large Language Models Using Multiprobe Approach
por: Amirizaniani, Maryam, et al.
Publicado: (2024)

Auditing Disability Representation in Vision-Language Models
por: Panda, Srikant, et al.
Publicado: (2026)

How Teachers Can Use Large Language Models and Bloom's Taxonomy to Create Educational Quizzes
por: Elkins, Sabina, et al.
Publicado: (2024)

When Choices Become Priors: Contrastive Decoding for Scientific Figure Multiple-Choice QA
por: Roh, Taeyun, et al.
Publicado: (2026)

When Chain-of-Thought Backfires: Evaluating Prompt Sensitivity in Medical Language Models
por: Sadanandan, Binesh, et al.
Publicado: (2026)

LLMs are Capable of Misaligned Behavior Under Explicit Prohibition and Surveillance
por: Ivanov, Igor
Publicado: (2025)

When Noise Lowers The Loss: Rethinking Likelihood-Based Evaluation in Music Large Language Models
por: Li, Xiaosha, et al.
Publicado: (2026)

When Benchmarks are Targets: Revealing the Sensitivity of Large Language Model Leaderboards
por: Alzahrani, Norah, et al.
Publicado: (2024)

Maximally Permissive Reward Machines
por: Varricchione, Giovanni, et al.
Publicado: (2024)

Style Attack Disguise: When Fonts Become a Camouflage for Adversarial Intent
por: Zhang, Yangshijie, et al.
Publicado: (2025)

AuditWen:An Open-Source Large Language Model for Audit
por: Huang, Jiajia, et al.
Publicado: (2024)

When Reasoning Traces Become Performative: Step-Level Evidence that Chain-of-Thought Is an Imperfect Oversight Channel
por: Li, Wenkai, et al.
Publicado: (2026)

Should AI Become an Intergenerational Civil Right?
por: Crowcroft, Jon, et al.
Publicado: (2025)

When the Prompt Becomes Visual: Vision-Centric Jailbreak Attacks for Large Image Editing Models
por: Hou, Jiacheng, et al.
Publicado: (2026)

TRUST: A Decentralized Framework for Auditing Large Language Model Reasoning
por: Huang, Morris Yu-Chao, et al.
Publicado: (2025)

Weak Permission is not Well-Founded, Grounded and Stable
por: Governatori, Guido
Publicado: (2024)

Privacy Auditing of Large Language Models
por: Panda, Ashwinee, et al.
Publicado: (2025)

Auditing Language Model Unlearning via Information Decomposition
por: Goel, Anmol, et al.
Publicado: (2026)

CALM: Curiosity-Driven Auditing for Large Language Models
por: Zheng, Xiang, et al.
Publicado: (2025)