Cover Image

Speech-Audio Compositional Attacks on Multimodal LLMs and Their Mitigation with SALMONN-Guard

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Yang, Yudong, Zhang, Xuezhen, Han, Zhifeng, Wang, Siyin, Zhuang, Jimin, Jin, Zengrui, Shao, Jing, Sun, Guangzhi, Zhang, Chao
Format:	Preprint
Published:	2025
Subjects:	Sound Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

video-SALMONN 2: Caption-Enhanced Audio-Visual Large Language Models
by: Tang, Changli, et al.
Published: (2025)

SALMONN-omni: A Codec-free LLM for Full-duplex Speech Understanding and Generation
by: Yu, Wenyi, et al.
Published: (2024)

Audio-Conditioned Diffusion LLMs for ASR and Deliberation Processing
by: Wang, Mengqi, et al.
Published: (2025)

Enabling Auditory Large Language Models for Automatic Speech Quality Evaluation
by: Wang, Siyin, et al.
Published: (2024)

MetaSICL: Adapting Audiroty LLM via Meta Speech In-Context Learning
by: Zheng, Haolong, et al.
Published: (2026)

SALMONN: Towards Generic Hearing Abilities for Large Language Models
by: Tang, Changli, et al.
Published: (2023)

video-SALMONN-o1: Reasoning-enhanced Audio-visual Large Language Model
by: Sun, Guangzhi, et al.
Published: (2025)

Optimising Neural Speech Codecs for 300bps Communication using Reinforcement Learning
by: Wang, Junyi, et al.
Published: (2026)

Extract and Diffuse: Latent Integration for Improved Diffusion-based Speech and Vocal Enhancement
by: Yang, Yudong, et al.
Published: (2024)

Audio Large Language Models Can Be Descriptive Speech Quality Evaluators
by: Chen, Chen, et al.
Published: (2025)

Protecting Bystander Privacy via Selective Hearing in Audio LLMs
by: Zhan, Xiao, et al.
Published: (2025)

ClariCodec: Optimising Neural Speech Codes for 200bps Communication using Reinforcement Learning
by: Wang, Junyi, et al.
Published: (2026)

video-SALMONN S: Memory-Enhanced Streaming Audio-Visual LLM
by: Sun, Guangzhi, et al.
Published: (2025)

Low-Rank and Sparse Model Merging for Multi-Lingual Speech Recognition and Translation
by: Zhao, Qiuming, et al.
Published: (2025)

SyncGuard: Robust Audio Watermarking Capable of Countering Desynchronization Attacks
by: Gan, Zhenliang, et al.
Published: (2025)

Bayesian Example Selection Improves In-Context Learning for Speech, Text, and Visual Modalities
by: Wang, Siyin, et al.
Published: (2024)

Investigation of Zero-shot Text-to-Speech Models for Enhancing Short-Utterance Speaker Verification
by: Zhao, Yiyang, et al.
Published: (2025)

Robust Zero-Shot Text-to-Speech Synthesis with Reverse Inference Optimization
by: Hu, Yuchen, et al.
Published: (2024)

Can Whisper perform speech-based in-context learning?
by: Wang, Siyin, et al.
Published: (2023)

SALMONN-omni: A Standalone Speech LLM without Codec Injection for Full-duplex Conversation
by: Yu, Wenyi, et al.
Published: (2025)

OCR-Enhanced Multimodal ASR Can Read While Listening
by: Chen, Junli, et al.
Published: (2026)

Joint Speaker Features Learning for Audio-visual Multichannel Speech Separation and Recognition
by: Li, Guinan, et al.
Published: (2024)

video-SALMONN: Speech-Enhanced Audio-Visual Large Language Models
by: Sun, Guangzhi, et al.
Published: (2024)

Codec-Robust Attacks on Audio LLMs
by: Roh, Jaechul, et al.
Published: (2026)

From Coarse to Fine: Recursive Audio-Visual Semantic Enhancement for Speech Separation
by: Xue, Ke, et al.
Published: (2025)

JAEGER: Joint 3D Audio-Visual Grounding and Reasoning in Simulated Physical Environments
by: Liu, Zhan, et al.
Published: (2026)

Multimodal Attention Merging for Improved Speech Recognition and Audio Event Classification
by: Sundar, Anirudh S., et al.
Published: (2023)

SpeechGuard: Exploring the Adversarial Robustness of Multimodal Large Language Models
by: Peri, Raghuveer, et al.
Published: (2024)

HoliAntiSpoof: Audio LLM for Holistic Speech Anti-Spoofing
by: Xu, Xuenan, et al.
Published: (2026)

Can Large Language Models Understand Spatial Audio?
by: Tang, Changli, et al.
Published: (2024)

Minimising Biasing Word Errors for Contextual ASR with the Tree-Constrained Pointer Generator
by: Sun, Guangzhi, et al.
Published: (2022)

Towards Explicit Acoustic Evidence Perception in Audio LLMs for Speech Deepfake Detection
by: Guo, Xiaoxuan, et al.
Published: (2026)

Ti-Audio: The First Multi-Dialectal End-to-End Speech LLM for Tibetan
by: Wang, Jialing, et al.
Published: (2026)

Mitigating Attention Sinks and Massive Activations in Audio-Visual Speech Recognition with LLMs
by: Anand, et al.
Published: (2025)

Can Large Audio Language Models Understand Audio Well? Speech, Scene and Events Understanding Benchmark for LALMs
by: Yin, Han, et al.
Published: (2025)

AudioFab: Building A General and Intelligent Audio Factory through Tool Learning
by: Zhu, Cheng, et al.
Published: (2025)

UTI-LLM: A Personalized Articulatory-Speech Therapy Assistance System Based on Multimodal Large Language Model
by: Yang, Yudong, et al.
Published: (2025)

Wav2Prompt: End-to-End Speech Prompt Generation and Tuning For LLM in Zero and Few-shot Learning
by: Deng, Keqi, et al.
Published: (2024)

Unfolding A Few Structures for The Many: Memory-Efficient Compression of Conformer and Speech Foundation Models
by: Li, Zhaoqing, et al.
Published: (2025)

SynthGuard: An Open Platform for Detecting AI-Generated Multimedia with Multimodal LLMs
by: Desai, Shail, et al.
Published: (2025)