Imagen de Portada

Red-teaming Activation Probes using Prompted LLMs

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Blandfort, Phil, Graham, Robert
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Direction-Flipped Influence Audits Reveal Hidden Structure in Moral Choices of LLMs
por: Blandfort, Phil, et al.
Publicado: (2026)

Active Attacks: Red-teaming LLMs via Adaptive Environments
por: Yun, Taeyoung, et al.
Publicado: (2025)

Detecting High-Stakes Interactions with Activation Probes
por: McKenzie, Alex, et al.
Publicado: (2025)

Curiosity-driven Red-teaming for Large Language Models
por: Hong, Zhang-Wei, et al.
Publicado: (2024)

Detection of adversarial intent in Human-AI teams using LLMs
por: Musaffar, Abed K., et al.
Publicado: (2026)

From Seed to Harvest: Augmenting Human Creativity with AI for Red-teaming Text-to-Image Models
por: Quaye, Jessica, et al.
Publicado: (2025)

Probe-Free Low-Rank Activation Intervention
por: Jiang, Chonghe, et al.
Publicado: (2025)

LeakAgent: RL-based Red-teaming Agent for LLM Privacy Leakage
por: Nie, Yuzhou, et al.
Publicado: (2024)

Probing Knowledge Holes in Unlearned LLMs
por: Ko, Myeongseob, et al.
Publicado: (2025)

APEX: Probing Neural Networks via Activation Perturbation
por: Ren, Tao, et al.
Publicado: (2026)

ContextBench: Modifying Contexts for Targeted Latent Activation
por: Graham, Robert, et al.
Publicado: (2025)

Beyond Naïve Prompting: Strategies for Improved Context-aided Forecasting with LLMs
por: Ashok, Arjun, et al.
Publicado: (2025)

Time-Prompt: Integrated Heterogeneous Prompts for Unlocking LLMs in Time Series Forecasting
por: Wang, Zesen, et al.
Publicado: (2025)

CAP: Controllable Alignment Prompting for Unlearning in LLMs
por: Wang, Zhaokun, et al.
Publicado: (2026)

Probing Graph Neural Network Activation Patterns Through Graph Topology
por: Tori, Floriano, et al.
Publicado: (2026)

Fast and Accurate Probing of In-Training LLMs' Downstream Performances
por: Liu, Zhichen, et al.
Publicado: (2026)

Activated LoRA: Fine-tuned LLMs for Intrinsics
por: Greenewald, Kristjan, et al.
Publicado: (2025)

Drag-and-Drop LLMs: Zero-Shot Prompt-to-Weights
por: Liang, Zhiyuan, et al.
Publicado: (2025)

Evolutionary System Prompt Learning for Reinforcement Learning in LLMs
por: Zhang, Lunjun, et al.
Publicado: (2026)

RAST: Reasoning Activation in LLMs via Small-model Transfer
por: Ouyang, Siru, et al.
Publicado: (2025)

Meta-Learning for Cold-Start Personalization in Prompt-Tuned LLMs
por: Zhao, Yushang, et al.
Publicado: (2025)

GRP-Obliteration: Unaligning LLMs With a Single Unlabeled Prompt
por: Russinovich, Mark, et al.
Publicado: (2026)

Enhancing Instruction Following of LLMs via Activation Steering with Dynamic Rejection
por: Kang, Minjae, et al.
Publicado: (2026)

A Fragile Number Sense: Probing the Elemental Limits of Numerical Reasoning in LLMs
por: Rahman, Roussel, et al.
Publicado: (2025)

The case for delegated AI autonomy for Human AI teaming in healthcare
por: Jia, Yan, et al.
Publicado: (2025)

ReLU$^2$ Wins: Discovering Efficient Activation Functions for Sparse LLMs
por: Zhang, Zhengyan, et al.
Publicado: (2024)

Prompted Policy Search: Reinforcement Learning through Linguistic and Numerical Reasoning in LLMs
por: Zhou, Yifan, et al.
Publicado: (2025)

GraphiT: Efficient Node Classification on Text-Attributed Graphs with Prompt Optimized LLMs
por: Khoshraftar, Shima, et al.
Publicado: (2025)

Can LLMs Effectively Leverage Graph Structural Information through Prompts, and Why?
por: Huang, Jin, et al.
Publicado: (2023)

Steering Frozen LLMs: Adaptive Social Alignment via Online Prompt Routing
por: Zhang, Zeyu, et al.
Publicado: (2026)

How Susceptible are LLMs to Influence in Prompts?
por: Anagnostidis, Sotiris, et al.
Publicado: (2024)

Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers
por: Karvonen, Adam, et al.
Publicado: (2025)

Steer Like the LLM: Activation Steering that Mimics Prompting
por: Heyman, Geert, et al.
Publicado: (2026)

Probe Pruning: Accelerating LLMs through Dynamic Pruning via Model-Probing
por: Le, Qi, et al.
Publicado: (2025)

Reinforcement Learning Fine-Tuning Enhances Activation Intensity and Diversity in the Internal Circuitry of LLMs
por: Zhang, Honglin, et al.
Publicado: (2025)

CoLA: Compute-Efficient Pre-Training of LLMs via Low-Rank Activation
por: Liu, Ziyue, et al.
Publicado: (2025)

ZorBA: Zeroth-order Federated Fine-tuning of LLMs with Heterogeneous Block Activation
por: Meng, Chuiyang, et al.
Publicado: (2026)

TrustGLM: Evaluating the Robustness of GraphLLMs Against Prompt, Text, and Structure Attacks
por: Zhang, Qihai, et al.
Publicado: (2025)

DGP: A Dual-Granularity Prompting Framework for Fraud Detection with Graph-Enhanced LLMs
por: Li, Yuan, et al.
Publicado: (2025)

Error-Driven Prompt Optimization for Arithmetic Reasoning
por: Pándy, Árpád, et al.
Publicado: (2025)