Image de couverture de livre

Signal in the Noise: Polysemantic Interference Transfers and Predicts Cross-Model Influence

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Gong, Bofan, Lai, Shiyang, Evans, James, Song, Dawn
Format:	Preprint
Publié:	2025
Sujets:	Artificial Intelligence Computation and Language Cryptography and Security
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

WebSentinel: Detecting and Localizing Prompt Injection Attacks for Web Agents
par: Wang, Xilong, et autres
Publié: (2026)

SafeKey: Amplifying Aha-Moment Insights for Safety Reasoning
par: Zhou, Kaiwen, et autres
Publié: (2025)

Measuring Real-World Prompt Injection Attacks in LLM-based Resume Screening
par: Zhang, Mohan, et autres
Publié: (2026)

When Benign Inputs Lead to Severe Harms: Eliciting Unsafe Unintended Behaviors of Computer-Use Agents
par: Jones, Jaylen, et autres
Publié: (2026)

DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models
par: Wang, Boxin, et autres
Publié: (2023)

RigorLLM: Resilient Guardrails for Large Language Models against Undesired Content
par: Yuan, Zhuowen, et autres
Publié: (2024)

CATMark: A Context-Aware Thresholding Framework for Robust Cross-Task Watermarking in Large Language Models
par: Zhang, Yu, et autres
Publié: (2025)

Have You Merged My Model? On The Robustness of Large Language Model IP Protection Methods Against Model Merging
par: Cong, Tianshuo, et autres
Publié: (2024)

MMDT: Decoding the Trustworthiness and Safety of Multimodal Foundation Models
par: Xu, Chejian, et autres
Publié: (2025)

Where to Start Alignment? Diffusion Large Language Model May Demand a Distinct Position
par: Xie, Zhixin, et autres
Publié: (2025)

JailbreakEval: An Integrated Toolkit for Evaluating Jailbreak Attempts Against Large Language Models
par: Ran, Delong, et autres
Publié: (2024)

LoopLLM: Transferable Energy-Latency Attacks in LLMs via Repetitive Generation
par: Li, Xingyu, et autres
Publié: (2025)

FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts
par: Gong, Yichen, et autres
Publié: (2023)

Resource Consumption Threats in Large Language Models
par: Zhang, Yuanhe, et autres
Publié: (2026)

SynGhost: Invisible and Universal Task-agnostic Backdoor Attack via Syntactic Transfer
par: Cheng, Pengzhou, et autres
Publié: (2024)

Breaking PEFT Limitations: Leveraging Weak-to-Strong Knowledge Transfer for Backdoor Attacks in LLMs
par: Zhao, Shuai, et autres
Publié: (2024)

Beyond the Tip of Efficiency: Uncovering the Submerged Threats of Jailbreak Attacks in Small Language Models
par: Yi, Sibo, et autres
Publié: (2025)

Effective and Efficient Jailbreaks of Black-Box LLMs with Cross-Behavior Attacks
par: Gohil, Vasudev
Publié: (2025)

No Attacker Needed: Unintentional Cross-User Contamination in Shared-State LLM Agents
par: Yang, Tiankai, et autres
Publié: (2026)

Omni-Safety under Cross-Modality Conflict: Vulnerabilities, Dynamics Mechanisms and Efficient Alignment
par: Wang, Kun, et autres
Publié: (2026)

How Jailbreak Defenses Work and Ensemble? A Mechanistic Investigation
par: Long, Zhuohang, et autres
Publié: (2025)

WAInjectBench: Benchmarking Prompt Injection Detections for Web Agents
par: Liu, Yinuo, et autres
Publié: (2025)

On Adversarial Robustness of Language Models in Transfer Learning
par: Turbal, Bohdan, et autres
Publié: (2024)

Fine-Tuning Language Models with Differential Privacy through Adaptive Noise Allocation
par: Li, Xianzhi, et autres
Publié: (2024)

Medical MLLM is Vulnerable: Cross-Modality Jailbreak and Mismatched Attacks on Medical Multimodal Large Language Models
par: Huang, Xijie, et autres
Publié: (2024)

Reverse-Engineering Model Editing on Language Models
par: Sun, Zhiyu, et autres
Publié: (2026)

Dagger Behind Smile: Fool LLMs with a Happy Ending Story
par: Song, Xurui, et autres
Publié: (2025)

Differentially Private Retrieval-Augmented Generation
par: Tang, Tingting, et autres
Publié: (2026)

Claim-Guided Textual Backdoor Attack for Practical Applications
par: Song, Minkyoo, et autres
Publié: (2024)

Learnable Linguistic Watermarks for Tracing Model Extraction Attacks on Large Language Models
par: Bai, Minhao, et autres
Publié: (2024)

sudo rm -rf agentic_security
par: Lee, Sejin, et autres
Publié: (2025)

Say Something Else: Rethinking Contextual Privacy as Information Sufficiency
par: Xiao, Yunze, et autres
Publié: (2026)

Tell me about yourself: LLMs are aware of their learned behaviors
par: Betley, Jan, et autres
Publié: (2025)

Exploring Backdoor Vulnerabilities of Chat Models
par: Hao, Yunzhuo, et autres
Publié: (2024)

Imperceptible Jailbreaking against Large Language Models
par: Gao, Kuofeng, et autres
Publié: (2025)

Large Reasoning Models Are Autonomous Jailbreak Agents
par: Hagendorff, Thilo, et autres
Publié: (2025)

Toward Cybersecurity-Expert Small Language Models
par: Levi, Matan, et autres
Publié: (2025)

Watermarking Language Models for Many Adaptive Users
par: Cohen, Aloni, et autres
Publié: (2024)

DP-MemArc: Differential Privacy Transfer Learning for Memory Efficient Language Models
par: Liu, Yanming, et autres
Publié: (2024)

Jailbreaking Large Language Models Through Content Concretization
par: Wahréus, Johan, et autres
Publié: (2025)