BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:DSS-GAN (Directional State Space GAN): A Generative Adversarial Ne
 twork with Directional State Spaces Utilizing the Mamba Architecture
DTSTART;VALUE=DATE-TIME:20260420T080000Z
DTEND;VALUE=DATE-TIME:20260420T090000Z
DTSTAMP;VALUE=DATE-TIME:20260415T232012Z
UID:indico-contribution-1947@events.ncbj.gov.pl
DESCRIPTION:Speakers: Aleksander Ogonowski ()\nDSS-GAN is the first genera
 tive adversarial network to utilize the Mamba architecture as a hierarchic
 al generator\, synthesizing images from noise under class-conditional sett
 ings. The key contribution is the Directional Latent Routing (DLR) mechani
 sm\, which decomposes the latent vector and class signal into direction-sp
 ecific sub-vectors\, each independently conditioning the corresponding sca
 nning direction along various spatial axes of the feature map.\n\nhttps://
 events.ncbj.gov.pl/event/480/contributions/1947/
LOCATION:Zoom https://cern.zoom.us/j/66151941204?pwd=n7upvvZYibexBhbtyn5kv
 Tpy36L0Wo.1
URL:https://events.ncbj.gov.pl/event/480/contributions/1947/
END:VEVENT
END:VCALENDAR