Diff-prior uses a diffusion model to learn an adaptive graph prior that performs denoising-style calibration on edge posteriors to yield more reliable interaction graphs in NRI tasks.
noise form
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
From Uniform to Learned Graph Priors: Diffusion for Structure Discovery
Diff-prior uses a diffusion model to learn an adaptive graph prior that performs denoising-style calibration on edge posteriors to yield more reliable interaction graphs in NRI tasks.