[KongchangAI]
Conceptscore prior / 分数函数

Score Prior

An estimate of the gradient of the logarithm of a probability density, pointing in the direction of steepest increase of probability density in the data distribution. It acts as a force field during denoising and is closely related to diffusion models and Tweedie's equation.

Timeline (last 90 days)

Oct 14

The core operation of the S³-PnP framework is to learn a score prior directly on the observation-driven intermediate states of ADMM, requiring neither clean training targets nor external pretrained weights.

Unverified50%

All Facts (1)

Unverified

The core operation of the S³-PnP framework is to learn a score prior directly on the observation-driven intermediate states of ADMM, requiring neither clean training targets nor external pretrained weights.

50%

Source Articles

S³-PnP: Self-Supervised Score Priors for Plug-and-Play Denoising in Scientific ImagingOct 14