Imagen de Portada

Representation Bending for Large Language Model Safety

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Yousefpour, Ashkan, Kim, Taeheon, Kwon, Ryan S., Lee, Seungbeen, Jeung, Wonje, Han, Seungju, Wan, Alvin, Ngan, Harrison, Yu, Youngjae, Choi, Jonghyun
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Computation and Language Cryptography and Security
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Large Language Models Still Exhibit Bias in Long Text
por: Jeung, Wonje, et al.
Publicado: (2024)

Adversarial Sample-Based Approach for Tighter Privacy Auditing in Final Model-Only Scenarios
por: Yoon, Sangyeon, et al.
Publicado: (2024)

Few-Shot Truly Benign DPO Attack for Jailbreaking LLMs
por: Yoon, Sangyeon, et al.
Publicado: (2026)

sudo rm -rf agentic_security
por: Lee, Sejin, et al.
Publicado: (2025)

An Information Theoretic Evaluation Metric For Strong Unlearning
por: Jeon, Dongjae, et al.
Publicado: (2024)

System-aware contextual digital twin for ICS anomaly diagnosis
por: Woo, Eungyu, et al.
Publicado: (2026)

Multi-Level Knowledge Distillation and Dynamic Self-Supervised Learning for Continual Learning
por: Kim, Taeheon, et al.
Publicado: (2025)

DataFreeShield: Defending Adversarial Attacks without Training Data
por: Lee, Hyeyoon, et al.
Publicado: (2024)

Safety and Security Analysis of Large Language Models: Benchmarking Risk Profile and Harm Potential
por: Akiri, Charankumar, et al.
Publicado: (2025)

Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders
por: Jiang, Laura, et al.
Publicado: (2026)

GraphAttack: Exploiting Representational Blindspots in LLM Safety Mechanisms
por: He, Sinan, et al.
Publicado: (2025)

Understanding and Mitigating Over-refusal for Large Language Models via Safety Representation
por: Zhang, Junbo, et al.
Publicado: (2025)

Fastrack: Fast IO for Secure ML using GPU TEEs
por: Wang, Yongqin, et al.
Publicado: (2024)

KDPrint: Passive Authentication using Keystroke Dynamics-to-Image Encoding via Standardization
por: Kim, Yooshin, et al.
Publicado: (2024)

SEAL: Entangled White-box Watermarks on Low-Rank Adaptation
por: Oh, Giyeong, et al.
Publicado: (2025)

Layer-Aware Representation Filtering: Purifying Finetuning Data to Preserve LLM Safety Alignment
por: Li, Hao, et al.
Publicado: (2025)

Unified Steganography via Implicit Neural Representation
por: Song, Qi, et al.
Publicado: (2025)

PeAR: A Static Binary Rewriting Framework for Binary-Only Fuzzing
por: Charles, Alvin, et al.
Publicado: (2026)

UniHENN: Designing Faster and More Versatile Homomorphic Encryption-based CNNs without im2col
por: Choi, Hyunmin, et al.
Publicado: (2024)

Characterization of GPU TEE Overheads in Distributed Data Parallel ML Training
por: Lee, Jonghyun, et al.
Publicado: (2025)

Towards Identification and Intervention of Safety-Critical Parameters in Large Language Models
por: Qi, Weiwei, et al.
Publicado: (2026)

Jailbreak Antidote: Runtime Safety-Utility Balance via Sparse Representation Adjustment in Large Language Models
por: Shen, Guobin, et al.
Publicado: (2024)

QUICstep: Evaluating connection migration based QUIC censorship circumvention
por: Lee, Seungju, et al.
Publicado: (2023)

SafeBench: A Safety Evaluation Framework for Multimodal Large Language Models
por: Ying, Zonghao, et al.
Publicado: (2024)

UAVCAN Dataset Description
por: Kim, Dongsung, et al.
Publicado: (2022)

Optimal partition selection with Rényi differential privacy
por: Harrison, Charlie, et al.
Publicado: (2026)

KUDA: Knowledge Unlearning by Deviating Representation for Large Language Models
por: Fang, Ce, et al.
Publicado: (2026)

CuFuzz: Hardening CUDA Programs through Transformation and Fuzzing
por: Singh, Saurabh, et al.
Publicado: (2026)

Cooking Up Risks: Benchmarking and Reducing Food Safety Risks in Large Language Models
por: Luo, Weidi, et al.
Publicado: (2026)

Beyond Fixed and Dynamic Prompts: Embedded Jailbreak Templates for Advancing LLM Security
por: Kim, Hajun, et al.
Publicado: (2025)

Blind-Touch: Homomorphic Encryption-Based Distributed Neural Network Inference for Privacy-Preserving Fingerprint Authentication
por: Choi, Hyunmin, et al.
Publicado: (2023)

Overthinking Loops in Agents: A Structural Risk via MCP Tools
por: Lee, Yohan, et al.
Publicado: (2026)

Permit: Permission-Aware Representation Intervention for Controlled Generation in Large Language Models
por: Sun, Pengcheng, et al.
Publicado: (2026)

Verifiable Exponential Mechanism for Median Estimation
por: Kwon, Hyukjun, et al.
Publicado: (2025)

Enforcing MAVLink Safety & Security Properties Via Refined Multiparty Session Types
por: Amorim, Arthur, et al.
Publicado: (2025)

GSPR: Aligning LLM Safeguards as Generalizable Safety Policy Reasoners
por: Li, Haoran, et al.
Publicado: (2025)

SoK: A Security Architect's View of Printed Circuit Board Attacks
por: Harrison, Jacob, et al.
Publicado: (2024)

Sublinear Risk-Limiting Audits from Direct Ballot Selection and Statistical Ballot Manifests
por: Fuller, Benjamin, et al.
Publicado: (2026)

Secure Coding with AI -- From Detection to Repair
por: Belozerov, Vladislav, et al.
Publicado: (2025)

SGuard-v1: Safety Guardrail for Large Language Models
por: Lee, JoonHo, et al.
Publicado: (2025)