Jae-Joon Kim

Multi-agent LoRA systems can slash time-to-first-token by 3.1× without sacrificing role specialization by decoupling shared base KV caches from precomputed low-rank adapter updates.

Chaining low-rank KV corrections directly across sequential agent workflows eliminates redundant prefills, slashing peak GPU memory by 3.7x and cutting time-to-first-token in half without sacrificing model accuracy.

Geometry-Change VLA (GC-VLA), which learns to predict multiview future-current geometry-change tokens from current observations and applies Geometry-Conditioned Residual Flow (GCRF), using a binary intervention router and a single bounded residual velocity policy learned from closed-loop feedback.

Dense matrix multiplication accelerators can surprisingly outperform dedicated sparse accelerators for sparse neural networks, offering better area and energy efficiency.