Image de couverture de livre

LatentGuard: Controllable Latent Steering for Robust Refusal of Attacks and Reliable Response Generation

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Shu, Huizhen, Li, Xuying, Li, Zhuo
Format:	Preprint
Publié:	2025
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

The Resurgence of GCG Adversarial Attacks on Large Language Models
par: Tan, Yuting, et autres
Publié: (2025)

Layer-Wise Perturbations via Sparse Autoencoders for Adversarial Text Generation
par: Shu, Huizhen, et autres
Publié: (2025)

Latent-space Attacks for Refusal Evasion in Language Models
par: Piras, Giorgio, et autres
Publié: (2026)

LatentRefusal: Latent-Signal Refusal for Unanswerable Text-to-SQL Queries
par: Ren, Xuancheng, et autres
Publié: (2026)

From Refusal Tokens to Refusal Control: Discovering and Steering Category-Specific Refusal Directions
par: Alagharu, Rishab, et autres
Publié: (2026)

Tracing the Dynamics of Refusal: Exploiting Latent Refusal Trajectories for Robust Jailbreak Detection
par: Hu, Xulin, et autres
Publié: (2026)

Exploring the Personality Traits of LLMs through Latent Features Steering
par: Yang, Shu, et autres
Publié: (2024)

Unveiling and Steering Connectome Organization with Interpretable Latent Variables
par: Li, Yubin, et autres
Publié: (2025)

Refusal Steering: Fine-grained Control over LLM Refusal Behaviour for Sensitive Topics
par: García-Ferrero, Iker, et autres
Publié: (2025)

Targeting the Core: A Simple and Effective Method to Attack RAG-based Agents via Direct LLM Manipulation
par: Li, Xuying, et autres
Publié: (2024)

Steer LLM Latents for Hallucination Detection
par: Park, Seongheon, et autres
Publié: (2025)

Latent Guard: a Safety Framework for Text-to-image Generation
par: Liu, Runtao, et autres
Publié: (2024)

RISER: Orchestrating Latent Reasoning Skills for Adaptive Activation Steering
par: Ye, Wencheng, et autres
Publié: (2026)

Programming Refusal with Conditional Activation Steering
par: Lee, Bruce W., et autres
Publié: (2024)

Feature-Guided SAE Steering for Refusal-Rate Control using Contrasting Prompts
par: Bhargav, Samaksh, et autres
Publié: (2025)

Controllable Mathematical Reasoning via Self-Optimizing Thought Vectors
par: LI, Xuying
Publié: (2025)

Latent Space Disentanglement via Activation Steering for Interpretable Attribute Control in Symbolic Music Generation
par: Prokopiou, Ioannis, et autres
Publié: (2026)

Transferable Latent-to-Latent Locomotion Policy for Efficient and Versatile Motion Control of Diverse Legged Robots
par: Zheng, Ziang, et autres
Publié: (2025)

AlphaSteer: Learning Refusal Steering with Principled Null-Space Constraint
par: Sheng, Leheng, et autres
Publié: (2025)

Latent Reward Steering: An Adaptive Inference-Time Framework that Implicitly Promotes Cognitive Behaviors in Reasoning LLMs
par: Li, Jiakang, et autres
Publié: (2026)

Output Length Effect on DeepSeek-R1's Safety in Forced Thinking
par: Li, Xuying, et autres
Publié: (2025)

Learn to Refuse: Making Large Language Models More Controllable and Reliable through Knowledge Scope Limitation and Refusal Mechanism
par: Cao, Lang
Publié: (2023)

Preemptive Detection and Steering of LLM Misalignment via Latent Reachability
par: Karnik, Sathwik, et autres
Publié: (2025)

On Effects of Steering Latent Representation for Large Language Model Unlearning
par: Huu-Tien, Dang, et autres
Publié: (2024)

Spatial-Aware Latent Initialization for Controllable Image Generation
par: Sun, Wenqiang, et autres
Publié: (2024)

RepIt: Steering Language Models with Concept-Specific Refusal Vectors
par: Siu, Vincent, et autres
Publié: (2025)

What Drives Representation Steering? A Mechanistic Case Study on Steering Refusal
par: Cheng, Stephen, et autres
Publié: (2026)

Latent Action Control for Reasoning-Guided Unified Image Generation
par: Zhai, Fuxiang, et autres
Publié: (2026)

Controllable and Stealthy Shilling Attacks via Dispersive Latent Diffusion
par: Qiao, Shutong, et autres
Publié: (2025)

Beyond a Single Direction: Chain-of-Thought Disrupts Simple Steering of Refusal
par: Yang, Kia-Jüng, et autres
Publié: (2026)

Latent Policy Steering with Embodiment-Agnostic Pretrained World Models
par: Wang, Yiqi, et autres
Publié: (2025)

Learning Latent Dynamic Robust Representations for World Models
par: Sun, Ruixiang, et autres
Publié: (2024)

Amortized Latent Steering: Low-Cost Alternative to Test-Time Optimization
par: Egbuna, Nathan, et autres
Publié: (2025)

Memory Inception: Latent-Space KV Cache Manipulation for Steering LLMs
par: Liu, Andy Zeyi, et autres
Publié: (2026)

FlowGuard: Towards Lightweight In-Generation Safety Detection for Diffusion Models via Linear Latent Decoding
par: Yang, Jinghan, et autres
Publié: (2026)

In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering
par: Liu, Sheng, et autres
Publié: (2023)

Beyond Explicit Refusals: Soft-Failure Attacks on Retrieval-Augmented Generation
par: Zhang, Wentao, et autres
Publié: (2026)

Steering to Say No: Configurable Refusal via Activation Steering in Vision Language Models
par: Yang, Jiaxi, et autres
Publié: (2026)

Precision Knowledge Editing: Enhancing Safety in Large Language Models
par: Li, Xuying, et autres
Publié: (2024)

Robust Latent Matters: Boosting Image Generation with Sampling Error Synthesis
par: Qiu, Kai, et autres
Publié: (2025)