Visual-affect-guided spatiotemporal graph modeling of disaster-image diffusion for embodied decision support

Xiao Zhang, Wei Zhang · 2026

Read Visual-affect-guided spatiotemporal graph modeling of disaster-image diffusion for embodied decision support by Xiao Zhang, Wei Zhang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13582661.txt.

Description

Xiao Zhang and Wei Zhang build, in a 2026 study, a model of how disaster images spread on social media, to aid robots that act in such events. Past tools read feeling at the whole-image level, or tracked spread with no split for feeling at all. Their model codes each image into a 128-part vector of visual feeling: force, fear, sadness, and bad tone. Those codes start the nodes of a time-aware graph net. Edges rest on true links, time order, and reach in space. Three heads then guess spread scale, spread speed, and how places rank.

Tests used 4,450 samples from a crisis image set, split in a set way. The full model hit strong scores on all three tasks. Parts cut one at a time show visual feeling, time steps, and space bounds each add a clear gain.

Catalog record

CREATORXiao Zhang, Wei Zhang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE47 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13582661

Browse the full library · About the Tower of Babel