Research map / Image generation / Fast diffusion sampling
Listening Through the Noise: Cauchy-Driven Diffusion Bridges for Robust Gastrointestinal Auscultation and Clinical Benchmarking
Dian Ding, Liren Dong, Yu Lu, Juntao Zhou, Ran Wang, Peng Li, Zhenyi Jia, Guangtao Xue
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 |
|---|---|---|---|
| replace Gaussian with Cauchy bridge noise (plain) (ablation) | DBIM · CLINBS | MAE: 16.49 -> 13.25 | Improved |
| replace Gaussian with Cauchy bridge noise (plain) (ablation) | DBIM · CLINBS | LSD: 25.21 -> 18.14 | Improved |
| replace Gaussian with Cauchy noise when L1 and Gram losses are used (ablation) | DBIM sampler · CLINBS | SSIM: 0.627 -> 0.706 | Improved |
| add L1 loss to Gaussian DBIM (ablation) | DBIM · CLINBS | MAE: 16.49 -> 12.47 | Improved |
| add L1 loss to Cauchy DBIM (ablation) | Cauchy DBIM · CLINBS | MAE: 13.25 -> 10.90 | Improved |
| add Gram loss to Cauchy DBIM (ablation) | Cauchy DBIM · CLINBS | SSIM: 0.594 -> 0.619 | Improved |
| add Gram loss to Gaussian DBIM (ablation) | DBIM · CLINBS | LPIPS: 0.296 -> 0.218 | Improved |
| add both L1 and Gram losses to Cauchy DBIM (ablation) | Cauchy DBIM · CLINBS | MAE: 13.25 -> 9.54 | Improved |
| use Gaussian scale-mixture sampler instead of DBIM (ablation) | Cauchy l1 gram · CLINBS | MAE: 9.54 -> 8.28 | Improved |
| use Gaussian scale-mixture sampler instead of DBIM (ablation) | Gaussian l1 gram · CLINBS | MAE: 10.52 -> 10.36 | Improved |
| share u across reverse steps instead of resampling (ablation) | Cauchy-DBIM · CLINBS | MAE: 8.75 -> 8.28 | Improved |
| add Gaussian noise on top of speech interference | Ours · CLINBS | MAE: 8.28 -> 9.10 | Hurt |
Closest papers by meaning
- PhysDiff: Physiology-based Dynamicity Disentangled Diffusion Model for Remote Physiological Measurement (AAAI 2025 · similarity 0.81)
- Region-Disentangled Diffusion Model for High-Fidelity PPG-to-ECG Translation (AAAI 2024 · similarity 0.81)
- Arctique: An artificial histopathological dataset unifying realism and controllability for uncertainty quantification (NeurIPS 2024 · similarity 0.81)
- EndoBench: A Comprehensive Evaluation of Multi-Modal Large Language Models for Endoscopy Analysis (NeurIPS 2025 · similarity 0.80)
- BGDB: Bernoulli-Gaussian Decision Block with Improved Denoising Diffusion Probabilistic Models (AAAI 2025 · similarity 0.80)
- CLIMB: Data Foundations for Large Scale Multimodal Clinical Foundation Models (ICML 2025 · similarity 0.80)
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.
