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-reasoner: Improving reasoning capability in large audio language models
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4representative citing papers
Semantic-level and verification-based uncertainty methods outperform token-level baselines for audio reasoning in ALLMs, but their relative performance on hallucination and unanswerable-question benchmarks is model- and task-dependent.
AVRT transfers reasoning to audio-visual models by distilling traces from single-modality teachers via LLM merger followed by SFT cold-start and RL, achieving SOTA on OmniBench, DailyOmni, and MMAR with 3B/7B models.
Step-Audio-R1.5 applies RLHF to audio reasoning models to maintain analytical performance while improving prosodic naturalness and immersion in extended spoken interactions.
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.
-
Walking Through Uncertainty: An Empirical Study of Uncertainty Estimation for Audio-Aware Large Language Models
Semantic-level and verification-based uncertainty methods outperform token-level baselines for audio reasoning in ALLMs, but their relative performance on hallucination and unanswerable-question benchmarks is model- and task-dependent.
-
AVRT: Audio-Visual Reasoning Transfer through Single-Modality Teachers
AVRT transfers reasoning to audio-visual models by distilling traces from single-modality teachers via LLM merger followed by SFT cold-start and RL, achieving SOTA on OmniBench, DailyOmni, and MMAR with 3B/7B models.
-
Step-Audio-R1.5 Technical Report
Step-Audio-R1.5 applies RLHF to audio reasoning models to maintain analytical performance while improving prosodic naturalness and immersion in extended spoken interactions.