Cover Image

Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Cyberey, Hannah, Evans, David
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Cryptography and Security Computers and Society
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

White-Box Sensitivity Auditing with Steering Vectors
by: Cyberey, Hannah, et al.
Published: (2026)

Unsupervised Concept Vector Extraction for Bias Control in LLMs
by: Cyberey, Hannah, et al.
Published: (2025)

Interpretable LLM Guardrails via Sparse Representation Steering
by: He, Zeqing, et al.
Published: (2025)

Analysing the Safety Pitfalls of Steering Vectors
by: Li, Yuxiao, et al.
Published: (2026)

Get my drift? Catching LLM Task Drift with Activation Deltas
by: Abdelnabi, Sahar, et al.
Published: (2024)

SecureForge: Finding and Preventing Vulnerabilities in LLM-Generated Code via Prompt Optimization
by: Liu, Houjun, et al.
Published: (2026)

Do Prevalent Bias Metrics Capture Allocational Harms from LLMs?
by: Cyberey, Hannah, et al.
Published: (2024)

SWAN: Semantic Watermarking with Abstract Meaning Representation
by: Ye, Ziping, et al.
Published: (2026)

AdaSteer: Your Aligned LLM is Inherently an Adaptive Jailbreak Defender
by: Zhao, Weixiang, et al.
Published: (2025)

Let's Measure the Elephant in the Room: Facilitating Personalized Automated Analysis of Privacy Policies at Scale
by: Zhao, Rui, et al.
Published: (2025)

ConVerse: Benchmarking Contextual Safety in Agent-to-Agent Conversations
by: Gomaa, Amr, et al.
Published: (2025)

An Investigation into Misuse of Java Security APIs by Large Language Models
by: Mousavi, Zahra, et al.
Published: (2024)

Domain-Independent Deception: A New Taxonomy and Linguistic Analysis
by: Verma, Rakesh M., et al.
Published: (2024)

Data Defenses Against Large Language Models
by: Agnew, William, et al.
Published: (2024)

How Susceptible are Large Language Models to Ideological Manipulation?
by: Chen, Kai, et al.
Published: (2024)

AI Agents May Always Fall for Prompt Injections
by: Abdelnabi, Sahar, et al.
Published: (2026)

LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet
by: Li, Nathaniel, et al.
Published: (2024)

Nuclear Deployed: Analyzing Catastrophic Risks in Decision-making of Autonomous LLM Agents
by: Xu, Rongwu, et al.
Published: (2025)

How Alignment and Jailbreak Work: Explain LLM Safety through Intermediate Hidden States
by: Zhou, Zhenhong, et al.
Published: (2024)

How Well Can LLM Agents Simulate End-User Security and Privacy Attitudes and Behaviors?
by: Li, Yuxuan, et al.
Published: (2026)

CoTGuard: Using Chain-of-Thought Triggering for Copyright Protection in Multi-Agent LLM Systems
by: Wen, Yan, et al.
Published: (2025)

Beyond Indistinguishability: Measuring Extraction Risk in LLM APIs
by: Liu, Ruixuan, et al.
Published: (2026)

Defending LLM Watermarking Against Spoofing Attacks with Contrastive Representation Learning
by: An, Li, et al.
Published: (2025)

An In-Depth Investigation of Data Collection in LLM App Ecosystems
by: Wu, Yuhao, et al.
Published: (2024)

Abusing the Internet of Medical Things: Evaluating Threat Models and Forensic Readiness for Multi-Vector Attacks on Connected Healthcare Devices
by: Straw, Isabel, et al.
Published: (2026)

Compiling Activation Steering into Weights via Null-Space Constraints for Stealthy Backdoors
by: Yin, Rui, et al.
Published: (2026)

On the Suitability of LLM-Driven Agents for Dark Pattern Audits
by: Sun, Chen, et al.
Published: (2026)

SALT: Steering Activations towards Leakage-free Thinking in Chain of Thought
by: Batra, Shourya, et al.
Published: (2025)

INR-Based Generative Steganography by Point Cloud Representation
by: Yangjie, Zhong, et al.
Published: (2024)

IsolateGPT: An Execution Isolation Architecture for LLM-Based Agentic Systems
by: Wu, Yuhao, et al.
Published: (2024)

LLM Novice Uplift on Dual-Use, In Silico Biology Tasks
by: Zhang, Chen Bo Calvin, et al.
Published: (2026)

Assessing LLM Response Quality in the Context of Technology-Facilitated Abuse
by: Prakash, Vijay, et al.
Published: (2026)

Autonomous Chain-of-Thought Distillation for Graph-Based Fraud Detection
by: Li, Yuan, et al.
Published: (2026)

Layer 2 Blockchains Simplified: A Survey of Vector Commitment Schemes, ZKP Frameworks, Layer-2 Data Structures and Verkle Trees
by: Kaur, Ekleen, et al.
Published: (2026)

Shortchanged: Uncovering and Analyzing Intimate Partner Financial Abuse in Consumer Complaints
by: Bhattacharya, Arkaprabha, et al.
Published: (2024)

Jailbreak-Tuning: Models Efficiently Learn Jailbreak Susceptibility
by: Murphy, Brendan, et al.
Published: (2025)

Phare: A Safety Probe for Large Language Models
by: Jeune, Pierre Le, et al.
Published: (2025)

Beyond Context: Large Language Models' Failure to Grasp Users' Intent
by: Hussain, Ahmed M., et al.
Published: (2025)

A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy
by: Wang, Huandong, et al.
Published: (2025)

SimMark: A Robust Sentence-Level Similarity-Based Watermarking Algorithm for Large Language Models
by: Dabiriaghdam, Amirhossein, et al.
Published: (2025)