Citation notice #7190 · 2026-07-11 11:51:00.114399+00:00

Understanding Latent Diffusability via Fisher Geometry

cites Information, which carries a correction notice dated 2020-10-21. One-hop deterministic notice: the citation edge exists in the Pith bibliography graph; no model judged whether the citation was load-bearing.

This is not a judgment on the citing paper.

Citing paper Cited paper on Pith Event page Original DOI Notice DOI File a formal challenge All reference changes

01Evidence

Raw extraction · citation context · bibliography index 1

Despite these empirical advances, the underlying causes of latent diffusion failure remain theoretically obscured. Crucially, the literature lacks a formal, quantitative framework for measuring how "diffusible" a latent space isbefore committing to the computationally expensive training of a full diffusion model. In this paper, we demystify latent-space diffusability through the rigorous lens of Fisher geometry [1], complementing recent information-geometric perspectives on diffusion trajectories [ 13]. We analyze diffusability viadenoising complexity, quantified by the rate of change of the Minimum Mean Squared Error (MMSE) along the continuous diffusion trajectory. We show that this denoising complexity fundamentally decomposes into two terms: the Fisher Information (FI) and its rate of dissipation, which we define as the Fisher Information Rate (FIR).

Parser render (TeX stripped for reading; raw above is the evidence)

Despite these empirical advances, the underlying causes of latent diffusion failure remain theoretically obscured. Crucially, the literature lacks a formal, quantitative framework for measuring how "diffusible" a latent space isbefore committing to the computationally expensive training of a full diffusion model. In this paper, we demystify latent-space diffusability through the rigorous lens of Fisher geometry [1], complementing recent information-geometric perspectives on diffusion trajectories [ 13]. We analyze diffusability viadenoising complexity, quantified by the rate of change of the Minimum Mean Squared Error (MMSE) along the continuous diffusion trajectory. We show that this denoising complexity fundamentally decomposes into two terms: the Fisher Information (FI) and its rate of dissipation, which we define as the Fisher Information Rate (FIR)

02Event

Type

Correction

Source

Crossref

Original DOI

10.1007/978-4-431-55978-8

Notice DOI

10.1007/978-4-431-55978-8_14

Date

2020-10-21

Title

Correction to: Information Geometry and Its Applications

Reasons

['Correction']

Work

Information (2016) Applied Mathematical Sciences

Schema constants (for re-runners): correction · crossref

03Dispute this notice

If this citation does not depend on the flagged claim, or the event is wrong, say so. Disputes are public. For a signed challenge against the paper itself, use the formal challenge form.