Subarna Tripathi

ByDeWay-V2 transforms spatial reasoning in MLLMs, achieving a 46% boost in accuracy while operating within strict resource constraints.

LAViFiT achieves superior recognition of surgical interactions by ensuring that vision encoders focus on action-relevant regions, overcoming the limitations of traditional fine-tuning methods.

Achieve object-level motion control in image-to-video generation without any training by cleverly exploiting attention maps and first-last-frame priors.