2 papers
cs.CV2026
Semantically Aligned Gradient-Driven Context-Preserving Image Editing
Chiranjeev Chiranjeev, Muskan Dosi, Mayank Vatsa +1
Instruction-guided image editing has a training-time blind spot. Generative editors are never required to semantically verify whether their outputs actually satisfy the instruction…
cs.CV2025
Harmonizing Geometry and Uncertainty: Diffusion with Hyperspheres
Muskan Dosi, Chiranjeev Chiranjeev, Kartik Thakral +2
Do contemporary diffusion models preserve the class geometry of hyperspherical data? Standard diffusion models rely on isotropic Gaussian noise in the forward process, inherently f…