English

DiffGAP: A Lightweight Diffusion Module in Contrastive Space for Bridging Cross-Model Gap

Computer Vision and Pattern Recognition 2025-03-18 v1 Artificial Intelligence Machine Learning Sound Audio and Speech Processing

Abstract

Recent works in cross-modal understanding and generation, notably through models like CLAP (Contrastive Language-Audio Pretraining) and CAVP (Contrastive Audio-Visual Pretraining), have significantly enhanced the alignment of text, video, and audio embeddings via a single contrastive loss. However, these methods often overlook the bidirectional interactions and inherent noises present in each modality, which can crucially impact the quality and efficacy of cross-modal integration. To address this limitation, we introduce DiffGAP, a novel approach incorporating a lightweight generative module within the contrastive space. Specifically, our DiffGAP employs a bidirectional diffusion process tailored to bridge the cross-modal gap more effectively. This involves a denoising process on text and video embeddings conditioned on audio embeddings and vice versa, thus facilitating a more nuanced and robust cross-modal interaction. Our experimental results on VGGSound and AudioCaps datasets demonstrate that DiffGAP significantly improves performance in video/text-audio generation and retrieval tasks, confirming its effectiveness in enhancing cross-modal understanding and generation capabilities.

Keywords

Cite

@article{arxiv.2503.12131,
  title  = {DiffGAP: A Lightweight Diffusion Module in Contrastive Space for Bridging Cross-Model Gap},
  author = {Shentong Mo and Zehua Chen and Fan Bao and Jun Zhu},
  journal= {arXiv preprint arXiv:2503.12131},
  year   = {2025}
}
R2 v1 2026-06-28T22:21:58.670Z