Yonggang Zhu
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
Sound×1AI×1NLP×1ML×1Audio and Speech Processing×1
Frequent co-authors
Research Timeline
2026
COMET: Concept Space Dissection of the Modality Gap in Audio-Text Multimodal Contrastive Embeddings
The paper introduces COMET, a novel PLS-SVD framework, to analyze the audio-text modality gap in CLAP models, showing that shared concepts are captured by a small subset of axes, and proposes a spectral truncation method to mitigate this gap for improved zero-shot performance.
Highlighted terms show continued research focus across papers