Tootfinder

'Russians are to blame' — Moldovans react to mass power outage after Moldova-Ukraine line malfunction: https://benborges.xyz/2026/01/31/russians-are-to-blame-moldovans.html

Heute gab's trotz dem am Mittwoch kaputt gegangenen G-Line mehrere freudige Dinge in Sachen #Fahrrad: Hab's nach mehreren Monaten Abstinenz wegen Terminkonflikten endlich mal wieder zur #CriticalMass #Zürich

(LinkedIn) Generative human intelligence for images anyone? https://www.linkedin.com/posts/petermeijer_the-voice-for-android-apps-on-google-play-activity-7413499467846500352-sfcJ

The vOICe for Android - Apps on Google Play | Peter Meijer
What is sensory substitution? When I speak to you and say, "Imagine a rising bright line", and you "see" a rising bright line in your mind's eye, that's mental imagery guided by sound, a form of sensory substitution. The vOICe vision BCI generalizes this such that it works for all images. And yes, being software-based, it is already in "mass production" well ahead of any Neuralink Blindsight brain implant, with well over 500,000 cumulative installs from Google Play https://lnkd.in/dGaF2br More…

Weighted Stochastic Differential Equation to Implement Wasserstein-Fisher-Rao Gradient Flow
Herlock Rahimi
https://arxiv.org/abs/2512.17878 https://arxiv.org/pdf/2512.17878 https://arxiv.org/html/2512.17878
arXiv:2512.17878v1 Announce Type: new
Abstract: Score-based diffusion models currently constitute the state of the art in continuous generative modeling. These methods are typically formulated via overdamped or underdamped Ornstein--Uhlenbeck-type stochastic differential equations, in which sampling is driven by a combination of deterministic drift and Brownian diffusion, resulting in continuous particle trajectories in the ambient space. While such dynamics enjoy exponential convergence guarantees for strongly log-concave target distributions, it is well known that their mixing rates deteriorate exponentially in the presence of nonconvex or multimodal landscapes, such as double-well potentials. Since many practical generative modeling tasks involve highly non-log-concave target distributions, considerable recent effort has been devoted to developing sampling schemes that improve exploration beyond classical diffusion dynamics.
A promising line of work leverages tools from information geometry to augment diffusion-based samplers with controlled mass reweighting mechanisms. This perspective leads naturally to Wasserstein--Fisher--Rao (WFR) geometries, which couple transport in the sample space with vertical (reaction) dynamics on the space of probability measures. In this work, we formulate such reweighting mechanisms through the introduction of explicit correction terms and show how they can be implemented via weighted stochastic differential equations using the Feynman--Kac representation. Our study provides a preliminary but rigorous investigation of WFR-based sampling dynamics, and aims to clarify their geometric and operator-theoretic structure as a foundation for future theoretical and algorithmic developments.
toXiv_bot_toot