3 papers
cs.CV2026
StruMPL: Multi-task Dense Regression under Disjoint Partial Supervision and MNAR Labels
Reza M. Asiyabi, Juan Alberto Molina-Valero, The SEOSAW Partnership +2
Estimating forest aboveground biomass (AGB) from Earth observation combines two structurally incompatible label sources: spaceborne lidar provides canopy structure at millions of l…
cs.LG2026
Process-Guided Concept Bottleneck Model
Reza M. Asiyabi, SEOSAW Partnership, Steven Hancock +1
Concept Bottleneck Models (CBMs) improve the explainability of black-box Deep Learning (DL) by introducing intermediate semantic concepts. However, standard CBMs often overlook dom…
econ.GN2026
Using satellite imagery to map rural marketplaces and monitor their activity at high frequency
Tillmann von Carnap, Reza M. Asiyabi, Paul Dingus +1
In many rural areas of low- and middle-income countries, weekly gatherings of buyers and sellers are the most tangible manifestation of the market economy. Knowing these markets' w…