Ukraine update from Giorgio Provincialli: Oleshky, Ukraine is the innermost circle of hell in Ukraine.
Please support independent journalism and support Ukraine by sharing the friend link to bypass the paywall: https://
#PhantastikPrompts 24.8.: Dein Roman wird in 5.000 Jahren bei einer archäologischen Ausgrabung gefunden. Was würden die Menschen daraus über deine Zeit und Welt schließen?
"Die Ansätze zum Bilden einer geschlechtergerechten deutschen Sprache, deren erste Versuche im 19. Jahrhundert wurzeln, wurden in den Werken von JTM konsequent weiterverfolgt. Da es sich bei den Erotika u…
Sphalerite (var schalenblende) from the Pomorzany Mine, Olkusz, Poland. A favorite desk specimen, given to me by a mine geologist there.
Exhibits typical concentric brown to yellow fine grained mineral layers, along with galena, pyrite. #MinCup26 #MineralMonday ⚒️
🇺🇦 #NowPlaying on BBCRadio3's #NightTracks
Olga Wojciechowska:
🎵 Images Imprisoned Within You
#OlgaWojciechowska
https://open.spotify.com/track/6MmFZGNiu5LCvuqegFkvn0
🇺🇦 #NowPlaying on KEXP's #MorningShow
IDLES:
🎵 Colossus
#IDLES
https://idlesband.bandcamp.com/track/colossus
https://open.spotify.com/track/4wf2tcIlvAxVBsjykZha71
Toward a mechanistic understanding of inference in visual cortex and diffusion models
Zeyu Yun, Alexander Belsten, Dasheng Bi, Zahra Kadkhodaie, Yubei Chen, Bruno A. Olshausen
https://arxiv.org/abs/2607.15693 https://arxiv.org/pdf/2607.15693 https://arxiv.org/html/2607.15693
arXiv:2607.15693v1 Announce Type: new
Abstract: We describe a model of perceptual inference in primary visual cortex (V1) equivalent to a minimal diffusion model whose function can be readily understood from its parameters. The model is based on sparse coding with a non-factorial prior over latent variables in the form of an unconstrained, pairwise interaction matrix, extending standard sparse coding inference to a general recurrent dynamical system. We efficiently train these recurrent dynamics using a denoising score-matching objective and implicit differentiation. After training on natural images, the learned interaction matrix mirrors the structure of horizontal connections in superficial layers of V1 that link neurons of similar orientation tuning. This model exhibits exceptionally good denoising performance, restoring image features such as extended contours amid extreme visual ambiguity, nearly matching the behavior of standard, black-box diffusion architectures in generalization regime. Owing to the model's simplicity, the network's Jacobian can be decomposed directly in terms of the interaction matrix between latent variables, revealing mechanistically how the recurrent dynamics assign high probability over a continuous family of natural structural deformations. Intriguingly, within this circuit, a large fraction of latent variables learn to disconnect from visual input altogether, essentially forming a hierarchical representation that appears to enforce global consistency among image features. Together, the model and results bridge two distinct domains: for neuroscience, it generates concrete, testable hypotheses regarding functional connectivity in recurrent neural circuits during perceptual inference tasks; for machine learning, it elucidates the internal mechanisms learned by diffusion models that allow them to generate infinitely many novel images from a finite training set.
toXiv_bot_toot
Heute vor 35 Jahren: Am 10. August 1991 führte eine geplante Wartung in einem #Kernkraftwerk #Tschernobyl zu einem Wasseraustritt aus dem Primärkreislauf. Ursache: Mögliche Sicherheitskulturmängel, Untersuchung läuft.
Luisier présente un budget pessimiste encore plus que les années précédents, en intégrant tous les imprévus possibles. Il ne sera pas pour autant plus précis. Au contraire, l’écart entre le budget et les comptes (plus positifs) va se creuser. Si on change les bases de calcul en pleine campagne, c’est donc une opération idéologique de communication. https://www.rts.ch/audio-podcast/2026/audio/forum-29357140.html