2 papers
cs.CV2026
DAGS: Disentangled Appearance-and-Geometry Steering of a Frozen Image DiT for Temporally Stabilized Generative Rendering
Karthik Mohan Kumar, Damian Andrysiak, Pedro Antonio Pena +2
Diffusion transformers (DiTs) generate high-fidelity images from text and image conditions, but their outputs carry large variance and their faithfulness to a desired target depend…
cs.AR2026
From Bit-Position Sensitivity to Unequal Error Protection for DNN Inference Memory
Muhammad Husnain Mubarik, Karthik Mohan Kumar, Pedro Antonio Pena +2
We characterize per-bit-position fault sensitivity in ML inference across 16 workloads -- spanning transformer-based models and attention-free CNNs -- and across three floating-poi…