On the Audio Hallucinations in Large Audio-Video Language Models
This addresses a specific issue for users of audio-video models, but it is incremental as it focuses on analyzing and classifying an existing problem rather than solving it fundamentally.
The paper tackles the problem of audio hallucinations in large audio-video language models, where models ignore audio content and generate descriptions based only on visual information, finding that 332 out of 1,000 sentences are hallucinated and achieving 87.9% F1 in classification with fine-tuning.
Large audio-video language models can generate descriptions for both video and audio. However, they sometimes ignore audio content, producing audio descriptions solely reliant on visual information. This paper refers to this as audio hallucinations and analyzes them in large audio-video language models. We gather 1,000 sentences by inquiring about audio information and annotate them whether they contain hallucinations. If a sentence is hallucinated, we also categorize the type of hallucination. The results reveal that 332 sentences are hallucinated with distinct trends observed in nouns and verbs for each hallucination type. Based on this, we tackle a task of audio hallucination classification using pre-trained audio-text models in the zero-shot and fine-tuning settings. Our experimental results reveal that the zero-shot models achieve higher performance (52.2% in F1) than the random (40.3%) and the fine-tuning models achieve 87.9%, outperforming the zero-shot models.