2 papers
cs.CL2026
CARVE: Content-Aware Recurrent with Value Efficiency for Chunk-Parallel Linear Attention
Sayak Dutta
Recurrent delta-rule models keep a fixed-size state matrix S (d_v x d_k) that compresses all past context. The state of the art (GDN-2) gates this update with element-wise matrix e…
cs.CV2025
UnCageNet: Tracking and Pose Estimation of Caged Animal
Sayak Dutta, Harish Katti, Shashikant Verma +1
Animal tracking and pose estimation systems, such as STEP (Simultaneous Tracking and Pose Estimation) and ViTPose, experience substantial performance drops when processing images a…