2026-05-29原文

Resolving Endpoint Underfitting in Diffusion Bridges via Noise Alignment

A new approach called Noise-Aligned Diffusion Bridge (NADB) addresses underfitting near the target endpoint in diffusion bridge models, improving image restoration and translation tasks.

Article intelligence

EngineersAdvanced

Key points

Current diffusion bridge models suffer from endpoint underfitting due to noise mismatch.
NADB introduces a mean network and noise-aligned mapping to correct this.
Experiments show improved performance in image restoration and translation.
Accepted at CVPR 2026, code is available.

Why it matters

This matters because current diffusion bridge models suffer from endpoint underfitting due to noise mismatch.

Technical impact

May affect model selection, inference cost, product capability, and evaluation benchmarks.

[2605.28962] Resolving Endpoint Underfitting in Diffusion Bridges via Noise Alignment

[Submitted on 27 May 2026]

Title:Resolving Endpoint Underfitting in Diffusion Bridges via Noise Alignment

View a PDF of the paper titled Resolving Endpoint Underfitting in Diffusion Bridges via Noise Alignment, by Yurong Gao and 4 other authors

View PDF HTML (experimental)

Abstract:Diffusion bridge models offer a powerful framework for connecting two data distributions, such as in image restoration and translation. Many existing methods learn this bridge by mimicking the score-matching formulation of standard diffusion models. In this work, we find that this way leads to an anomalous underfitting phenomenon near the target endpoint, as the process approaches the target distribution ($t \to 0$). This underfitting, characterized by significant drift in the predicted variance and direction, results from an excessively large discrepancy in noise levels between the network's input and its regression this http URL resolve this issue, we propose the Noise-Aligned Diffusion Bridge (NADB).Our approach reformulates the diffusion bridge by first employing a mean network to provide a cleaner conditional target, and then introducing a novel, noise-aligned mapping relationship. This new formulation resolves the noise mismatch and corrects the underfitting near the target endpoint. Experimental validation across multiple image restoration and image translation tasks demonstrates the effectiveness of our approach. Code is available at this https URL.

Comments: Accepted by CVPR2026

Subjects:

Computer Vision and Pattern Recognition (cs.CV)

Cite as: arXiv:2605.28962 [cs.CV]

(or arXiv:2605.28962v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2605.28962

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Yurong Gao [view email] [v1] Wed, 27 May 2026 18:07:21 UTC (7,768 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Resolving Endpoint Underfitting in Diffusion Bridges via Noise Alignment, by Yurong Gao and 4 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CV

new | recent | 2026-05

Change to browse by:

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)