Internal Safety Collapse in Frontier Large Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Wu, Yutao, Liu, Xiao, Gao, Yifeng, Zheng, Xiang, Huang, Hanxun, Li, Yige, Wang, Cong, Li, Bo, Ma, Xingjun, Jiang, Yu-Gang
Format:	Preprint
Published:	2026
Subjects:	Computation and Language Artificial Intelligence Cryptography and Security
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Shortcuts Everywhere and Nowhere: Exploring Multi-Trigger Backdoor Attacks
by: Li, Yige, et al.
Published: (2024)

ML-Bench&Guard: Policy-Grounded Multilingual Safety Benchmark and Guardrail for Large Language Models
by: Zhao, Yunhan, et al.
Published: (2026)

T2UE: Generating Unlearnable Examples from Text Descriptions
by: Ma, Xingjun, et al.
Published: (2025)

AutoBackdoor: Automating Backdoor Attacks via LLM Agents
by: Li, Yige, et al.
Published: (2025)

Backdoor4Good: Benchmarking Beneficial Uses of Backdoors in LLMs
by: Li, Yige, et al.
Published: (2026)

GenBreak: Red Teaming Text-to-Image Generators Using Large Language Models
by: Wang, Zilong, et al.
Published: (2025)

ADMIT: Few-shot Knowledge Poisoning Attacks on RAG-based Fact Checking
by: Wu, Yutao, et al.
Published: (2025)

AttackVLA: Benchmarking Adversarial and Backdoor Attacks on Vision-Language-Action Models
by: Li, Jiayu, et al.
Published: (2025)

Defining Cost Function of Steganography with Large Language Models
by: Wu, Hanzhou, et al.
Published: (2025)

End-to-End Anti-Backdoor Learning on Images and Time Series
by: Jiang, Yujing, et al.
Published: (2024)

SafeRedirect: Defeating Internal Safety Collapse via Task-Completion Redirection in Frontier LLMs
by: Pan, Chao, et al.
Published: (2026)

DropVLA: An Action-Level Backdoor Attack on Vision-Language-Action Models
by: Xu, Zonghuan, et al.
Published: (2025)

Just Ask: Curious Code Agents Reveal System Prompts in Frontier LLMs
by: Zheng, Xiang, et al.
Published: (2026)

Towards Million-Scale Adversarial Robustness Evaluation With Stronger Individual Attacks
by: Xie, Yong, et al.
Published: (2024)

Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses
by: Li, Xiao, et al.
Published: (2026)

Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety
by: Ma, Xingjun, et al.
Published: (2025)

BraveGuard: From Open-World Threats to Safer Computer-Use Agents
by: Feng, Yunhao, et al.
Published: (2026)

Q-MLLM: Vector Quantization for Robust Multimodal Large Language Model Security
by: Zhao, Wei, et al.
Published: (2025)

EnJa: Ensemble Jailbreak on Large Language Models
by: Zhang, Jiahao, et al.
Published: (2024)

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

Backdoor Token Unlearning: Exposing and Defending Backdoors in Pretrained Language Models
by: Jiang, Peihai, et al.
Published: (2025)

LeakyCLIP: Extracting Training Data from CLIP
by: Chen, Yunhao, et al.
Published: (2025)

DarkLLM: Learning Language-Driven Adversarial Attacks with Large Language Models
by: Sun, Ye, et al.
Published: (2026)

Unleashing the Unseen: Harnessing Benign Datasets for Jailbreaking Large Language Models
by: Zhao, Wei, et al.
Published: (2024)

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

Imperceptible Jailbreaking against Large Language Models
by: Gao, Kuofeng, et al.
Published: (2025)

EnchTable: Unified Safety Alignment Transfer in Fine-tuned Large Language Models
by: Wu, Jialin, et al.
Published: (2025)

CTRAP: Embedding Collapse Trap to Safeguard Large Language Models from Harmful Fine-Tuning
by: Yi, Biao, et al.
Published: (2025)

Blockchain Data Analysis in the Era of Large-Language Models
by: Toyoda, Kentaroh, et al.
Published: (2024)

BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models
by: Xiang, Zhen, et al.
Published: (2024)

TokenMark: A Modality-Agnostic Watermark for Pre-trained Transformers
by: Xu, Hengyuan, et al.
Published: (2024)

HarmRLVR: Weaponizing Verifiable Rewards for Harmful LLM Alignment
by: Liu, Yuexiao, et al.
Published: (2025)

Beyond Content Safety: Real-Time Monitoring for Reasoning Vulnerabilities in Large Language Models
by: Wang, Xunguang, et al.
Published: (2026)

Defenses at Odds: Measuring and Explaining Defense Conflicts in Large Language Models
by: Meng, Xiangtao, et al.
Published: (2026)

Practical Reasoning Interruption Attacks on Reasoning Large Language Models
by: Cui, Yu, et al.
Published: (2025)

Semantic Sleuth: Identifying Ponzi Contracts via Large Language Models
by: Wu, Cong, et al.
Published: (2024)

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

Safety Layers in Aligned Large Language Models: The Key to LLM Security
by: Li, Shen, et al.
Published: (2024)

StructuralSleight: Automated Jailbreak Attacks on Large Language Models Utilizing Uncommon Text-Organization Structures
by: Li, Bangxin, et al.
Published: (2024)

CPA-RAG:Covert Poisoning Attacks on Retrieval-Augmented Generation in Large Language Models
by: Li, Chunyang, et al.
Published: (2025)