提出结构化扩散桥模型，实现少配对数据下的高效跨模态转换。
Structured Diffusion Bridges: Inductive Bias for Denoising Diffusion Bridges
- 用对齐约束限制解空间，将成对数据转为可选辅助信息。
- 在无配对、半配对和全配对场景中均表现稳定，接近全配对性能。
- 适合数据标注成本高或配对困难的跨模态任务应用。
跨模态翻译本质上是欠约束问题，因为多个跨模态映射可能产生相同的边缘分布。近期研究显示扩散桥在该任务上有效。然而，大多数现有方法依赖完全配对的数据集，从而施加单一数据驱动约束。本文提出一种扩散桥框架，通过表征可接受解的空间并引入对齐约束来加以限制，将成对监督视为可选启发而非必要前提。我们在合成与真实数据上的跨模态翻译基准测试中验证了该方法，在无配对、半配对和全配对三种情形下均表现一致。显著的是，该方法在大幅放松配对要求的情况下仍能达到接近全配对的性能，并在无配对情形下依然有效。这些结果表明，扩散桥可作为超越全配对数据的灵活跨模态翻译基础。
原文摘要 · Abstract (English)
Modality translation is inherently under-constrained, as multiple cross-modal mappings may yield the same marginals. Recent work has shown that diffusion bridges are effective for this task. However, most existing approaches rely on fully paired datasets, thereby imposing a single data-driven constraint. We propose a diffusion-bridge framework that characterizes the space of admissible solutions and restricts it via alignment constraints, treating paired supervision as an optional heuristic rather than a prerequisite. We validate our method on synthetic and real modality translation benchmarks across unpaired, semi-paired, and paired regimes, showing consistent performance across supervision levels. Notably, \textbf{it achieves near fully-paired quality with a substantial relaxation in pairing requirements, and remaining applicable in the unpaired regime}. These results highlight diffusion bridges as a flexible foundation for modality translation beyond fully paired data.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。