Benign fine-tuning on audio data breaks safety alignment in Audio LLMs by raising jailbreak success rates up to 87%, with the dominant risk axis depending on model architecture and embedding proximity to harmful content.
Audio-language models for audio-centric tasks: A survey
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
citation-role summary
background 3
citation-polarity summary
years
2026 12roles
background 3polarities
background 3representative citing papers
Introduces the first benchmark for over-refusal in large audio language models using 3,000 pseudo-harmful audio samples and evaluates 12 models across six families, finding widespread over-refusal.
PairAlign learns compact variable-length token sequences for audio via self-alignment on paired content-preserving views, achieving 55% fewer archive tokens than VQ while preserving edit-distance retrieval at 12.71 tokens/s.
MSU-Bench is a new two-tier benchmark covering speaker grounding to dialogue reasoning in multi-speaker conversations, with Gemini-assisted annotation and human verification.
ELSA introduces an event-level semantic alignment metric for reference-free text-to-audio evaluation that reports higher correlation with human ratings than CLAP-based baselines across four benchmarks.
Acquisition route affects forgetting rates in multimodal models, with text-pathway knowledge forgetting faster than audio-pathway knowledge in music understanding tasks.
A wait-think-answer controller for LALMs is trained via SFT followed by six-reward DAPO, raising row-weighted accuracy from 67.6% to 70.3% and cutting post-endpoint thinking length by 14% on synthetic spoken QA while remaining functional on real recorded audio.
GlobeAudio is a new multilingual multicultural benchmark for naturalistic evaluation of large audio-language models, showing performance gaps especially for open-source models and low-resource languages.
A survey of Large Audio Language Models that establishes a taxonomy of trustworthiness vulnerabilities and proposes a Defense-in-Depth roadmap for audio intelligence.
Aligned acoustic concept tokens from eGeMAPS improve UAR in ALM-based SER on FAU-Aibo and IEMOCAP while shuffled or corrupted tokens reduce performance without collapsing predictions, indicating partial anchoring to audio.
Audio language models are benchmarked on five semantic and paralinguistic reasoning tasks to reveal limitations in handling spoken audio evidence, accent variation, and domain shifts.
A survey that provides a unified formulation of audio reasoning and reviews advances across Audio-to-Text, Audio-to-Speech, Audio-Visual, and Agentic paradigms while discussing challenges and future directions.
citing papers explorer
-
Benign Fine-Tuning Breaks Safety Alignment in Audio LLMs
Benign fine-tuning on audio data breaks safety alignment in Audio LLMs by raising jailbreak success rates up to 87%, with the dominant risk axis depending on model architecture and embedding proximity to harmful content.
-
AOR-Bench: Do Large Audio Language Models Over-Refuse Pseudo-Harmful Queries?
Introduces the first benchmark for over-refusal in large audio language models using 3,000 pseudo-harmful audio samples and evaluates 12 models across six families, finding widespread over-refusal.
-
PairAlign: A Framework for Sequence Tokenization via Self-Alignment with Applications to Audio Tokenization
PairAlign learns compact variable-length token sequences for audio via self-alignment on paired content-preserving views, achieving 55% fewer archive tokens than VQ while preserving edit-distance retrieval at 12.71 tokens/s.
-
MSU-Bench: Towards Speaker-Centric Understanding in Conversational Multi-Speaker Scenarios
MSU-Bench is a new two-tier benchmark covering speaker grounding to dialogue reasoning in multi-speaker conversations, with Gemini-assisted annotation and human verification.
-
ELSA: Acoustic Event-Level Semantic Alignment for Fine-Grained Reference-Free Text-to-Audio Evaluation
ELSA introduces an event-level semantic alignment metric for reference-free text-to-audio evaluation that reports higher correlation with human ratings than CLAP-based baselines across four benchmarks.
-
When the Same Musical Knowledge Forgets Differently: A Clean Probe of Pathway-Dependent Forgetting
Acquisition route affects forgetting rates in multimodal models, with text-pathway knowledge forgetting faster than audio-pathway knowledge in music understanding tasks.
-
Learning When to Think While Listening in Large Audio-Language Models
A wait-think-answer controller for LALMs is trained via SFT followed by six-reward DAPO, raising row-weighted accuracy from 67.6% to 70.3% and cutting post-endpoint thinking length by 14% on synthetic spoken QA while remaining functional on real recorded audio.
-
GlobeAudio: A Multilingual Multicultural Benchmark for Naturalistic Evaluation of Large Audio-Language Models
GlobeAudio is a new multilingual multicultural benchmark for naturalistic evaluation of large audio-language models, showing performance gaps especially for open-source models and low-resource languages.
-
A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook
A survey of Large Audio Language Models that establishes a taxonomy of trustworthiness vulnerabilities and proposes a Defense-in-Depth roadmap for audio intelligence.
-
Acoustic Cue Alignment in Audio Language Models for Speech Emotion Recognition
Aligned acoustic concept tokens from eGeMAPS improve UAR in ALM-based SER on FAU-Aibo and IEMOCAP while shuffled or corrupted tokens reduce performance without collapsing predictions, indicating partial anchoring to audio.
-
Afrispeech Semantics: Evaluating Audio Semantic Reasoning in Spoken Language Models Across Domains and Accents
Audio language models are benchmarked on five semantic and paralinguistic reasoning tasks to reveal limitations in handling spoken audio evidence, accent variation, and domain shifts.
-
A Survey of Audio Reasoning in Multimodal Foundation Models
A survey that provides a unified formulation of audio reasoning and reviews advances across Audio-to-Text, Audio-to-Speech, Audio-Visual, and Agentic paradigms while discussing challenges and future directions.