Image de couverture de livre

Sirens' Whisper: Inaudible Near-Ultrasonic Jailbreaks of Speech-Driven LLMs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Ling, Zijian, Hu, Pingyi, Gao, Xiuyong, Ma, Xiaojing, Zhou, Man, Feng, Jun, Lu, Songfeng, Zhang, Dongmei, Zhu, Bin Benjamin
Format:	Preprint
Publié:	2026
Sujets:	Cryptography and Security Artificial Intelligence Sound
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Whispering Under the Eaves: Protecting User Privacy Against Commercial and LLM-powered Automatic Speech Recognition Systems
par: Jin, Weifei, et autres
Publié: (2025)

Acoustic Interference: A New Paradigm Weaponizing Acoustic Latent Semantic for Universal Jailbreak against Large Audio Language Models
par: Wang, Yanyun, et autres
Publié: (2026)

Hybrid Audio Detection Using Fine-Tuned Audio Spectrogram Transformers: A Dataset-Driven Evaluation of Mixed AI-Human Speech
par: Huang, Kunyang, et autres
Publié: (2025)

Selective Masking Adversarial Attack on Automatic Speech Recognition Systems
par: Fang, Zheng, et autres
Publié: (2025)

HVAC-EAR: Eavesdropping Human Speech Using HVAC Systems
par: Tamiti, Tarikul Islam, et autres
Publié: (2025)

Vulnerabilities of Audio-Based Biometric Authentication Systems Against Deepfake Speech Synthesis
par: Hong, Mengze, et autres
Publié: (2026)

Hidden in Plain Sound: Environmental Backdoor Poisoning Attacks on Whisper, and Mitigations
par: Bartolini, Jonatan, et autres
Publié: (2024)

Benign Fine-Tuning Breaks Safety Alignment in Audio LLMs
par: Roh, Jaechul, et autres
Publié: (2026)

MelShield: Robust Mel-Domain Audio Watermarking for Provenance Attribution of AI Generated Synthesized Speech
par: Jin, Yutong, et autres
Publié: (2026)

When Fine-Tuning is Not Enough: Lessons from HSAD on Hybrid and Adversarial Audio Spoof Detection
par: Hu, Bin, et autres
Publié: (2025)

Multilingual and Multi-Accent Jailbreaking of Audio LLMs
par: Roh, Jaechul, et autres
Publié: (2025)

Whisper Smarter, not Harder: Adversarial Attack on Partial Suppression
par: Wong, Zheng Jie, et autres
Publié: (2025)

Can DeepFake Speech be Reliably Detected?
par: Liu, Hongbin, et autres
Publié: (2024)

SCDF: A Speaker Characteristics DeepFake Speech Dataset for Bias Analysis
par: Staněk, Vojtěch, et autres
Publié: (2025)

Decoding Deception: Understanding Automatic Speech Recognition Vulnerabilities in Evasion and Poisoning Attacks
par: G, Aravindhan, et autres
Publié: (2025)

Privacy in Speech Technology
par: Bäckström, Tom
Publié: (2023)

Smark: A Watermark for Text-to-Speech Diffusion Models via Discrete Wavelet Transform
par: Zhang, Yichuan, et autres
Publié: (2025)

SOLIDO: A Robust Watermarking Method for Speech Synthesis via Low-Rank Adaptation
par: Li, Yue, et autres
Publié: (2025)

Beyond Content: A Comprehensive Speech Toxicity Dataset and Detection Framework Incorporating Paralinguistic Cues
par: Ba, Zhongjie, et autres
Publié: (2026)

Where Do Backdoors Live? A Component-Level Analysis of Backdoor Propagation in Speech Language Models
par: Fortier, Alexandrine, et autres
Publié: (2025)

MerkleSpeech: Public-Key Verifiable, Chunk-Localised Speech Provenance via Perceptual Fingerprints and Merkle Commitments
par: Ono, Tatsunori
Publié: (2026)

Invisible Ears at Your Fingertips: Acoustic Eavesdropping via Mouse Sensors
par: Fakih, Mohamad, et autres
Publié: (2025)

DECKER: Domain-invariant Embedding for Cross-Keyboard Extraction and Recognition
par: Maurya, Bikrant Bikram Pratap, et autres
Publié: (2026)

Mirage Fools the Ear, Mute Hides the Truth: Precise Targeted Adversarial Attacks on Polyphonic Sound Event Detection Systems
par: Su, Junjie, et autres
Publié: (2025)

Audio Pirates: Black-box Audio Watermark Removal via Diffusion Priors
par: Yao, Lingfeng, et autres
Publié: (2026)

ClearMask: Noise-Free and Naturalness-Preserving Protection Against Voice Deepfake Attacks
par: Wang, Yuanda, et autres
Publié: (2025)

DASM: Domain-Aware Sharpness Minimization for Multi-Domain Voice Stream Steganalysis
par: Zhou, Pengcheng, et autres
Publié: (2026)

SARSteer: Safeguarding Large Audio-Language Models via Safe-Ablated Refusal Steering
par: Lin, Weilin, et autres
Publié: (2025)

Lightweight Protection for Privacy in Offloaded Speech Understanding
par: Cai, Dongqi
Publié: (2024)

Adversarial Attacks and Defenses for Speech Recognition Systems
par: Żelasko, Piotr, et autres
Publié: (2021)

JALMBench: Benchmarking Jailbreak Vulnerabilities in Audio Language Models
par: Peng, Zifan, et autres
Publié: (2025)

Frame-level Temporal Difference Learning for Partial Deepfake Speech Detection
par: Li, Menglu, et autres
Publié: (2025)

SafeSpeech: Robust and Universal Voice Protection Against Malicious Speech Synthesis
par: Zhang, Zhisheng, et autres
Publié: (2025)

WaLi: Can Pressure Sensors in HVAC Systems Capture Human Speech?
par: Tamiti, Tarikul Islam, et autres
Publié: (2025)

LENS-DF: Deepfake Detection and Temporal Localization for Long-Form Noisy Speech
par: Liu, Xuechen, et autres
Publié: (2025)

Why Speech Deepfake Detectors Won't Generalize: The Limits of Detection in an Open World
par: Berisha, Visar, et autres
Publié: (2025)

Zero-Query Adversarial Attack on Black-box Automatic Speech Recognition Systems
par: Fang, Zheng, et autres
Publié: (2024)

SpeechVerifier: Robust Acoustic Fingerprint against Tampering Attacks via Watermarking
par: Yao, Lingfeng, et autres
Publié: (2025)

A Portable and Stealthy Inaudible Voice Attack Based on Acoustic Metamaterials
par: Ning, Zhiyuan, et autres
Publié: (2025)

Cross-Technology Generalization in Synthesized Speech Detection: Evaluating AST Models with Modern Voice Generators
par: Ustinov, Andrew, et autres
Publié: (2025)