Research map / Image generation / Fast diffusion sampling
Frequency-Dependent Scheduled Schrödinger Bridge for Underwater Acoustic Signal Denoising
Pengsen Zhu, Lina Gao, Yulong Huang, Lifeng Liu, Zeru Yang, Yonggang Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add frequency-dependent scheduling (FDS) (ablation) | SB with VE schedule · ShipsEar -5 dB | SI-SNRi: 9.6596 -> 10.4751 | Improved |
| add FDS (ablation) | SB with VP schedule · ShipsEar -5 dB | SI-SNRi: 8.0616 -> 8.4488 | Improved |
| add FDS (ablation) | SB with SubVP schedule · ShipsEar -5 dB | SI-SNRi: 1.5446 -> 10.4051 | Improved |
| add FDS to FDSSB default schedule (ablation) | FDSSB · ShipsEar -5 dB | SI-SNRi: 11.6437 -> 13.4391 | Improved |
| set modulation intensity tau=1.0 vs other values (ablation) | FDSSB · ShipsEar -5 dB | SI-SNRi: 11.4971 (tau 0.3) -> 13.4391 (tau 1.0) | Improved |
| Schrodinger-bridge with FDS vs strongest baseline RGDNet | FDSSB · ShipsEar -5 dB | SI-SNRi: 10.2569 -> 13.4391 | Improved |
| FDSSB vs RGDNet at -10 dB | FDSSB · ShipsEar -10 dB (mixed) | SI-SNRi: 13.9406 -> 17.8364 | Improved |
| FDSSB vs RGDNet on -10 dB average | FDSSB · ShipsEar -10 dB (average) | SI-SNRi: 13.0362 -> 13.3168 (small margin vs mixed 13.9406 -> 17.8364) | Improved |
Closest papers by meaning
- Finding Local Diffusion Schrodinger Bridge using Kolmogorov-Arnold Network (CVPR 2025 · similarity 0.85)
- SBNO : Schrödinger Bridge Neural Operator for the Forward and Inverse Problems under Missing Information (NeurIPS 2026 · similarity 0.85)
- Improved sampling via learned diffusions (ICLR 2024 · similarity 0.84)
- Adjoint Schrödinger Bridge Sampler (NeurIPS 2025 · similarity 0.84)
- LENS: Low-Frequency Eigen Noise Shaping for Efficient Diffusion Sampling (NeurIPS 2026 · similarity 0.83)
- Dequantified Diffusion-Schrödinger Bridge for Density Ratio Estimation (ICML 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
