Raahim Arbaz
Building computer vision for operating rooms and production LLM software.
Now: CVPR 2027 paper in preparation
Building computer vision for operating rooms and production LLM software.
Now: CVPR 2027 paper in preparation
+3 earlier
Founding engineer · CV & ML
@infiniV · Lahore, Pakistan
Founding engineer at Gray Scrubs AI. OR computer vision and LLM software.
--:-- · PKT UTC+5
1,403 contributions in the last year
07 / RESEARCH
Interpretability of robot vision-language-action models, and satellite imagery for climate.
Interpretability of robot vision-language-action models: model stitching, probes and sparse crosscoders in pre-registered experiments on sealed test sets.
Pollution-source detection (factory and brick-kiln chimneys) in South Asian cities from satellite imagery: a Vision Transformer filters high-resolution tiles, then Remote CLIP refines detection in two passes.