Jiayu Chen

Achieve professional-grade video mashups by mimicking a human production pipeline, using hierarchical agents to handle global structure, editing intent, and fine-grained shot selection.

Ditch the training: SVOO achieves up to 1.93x speedup in video generation with sparse attention by exploiting the intrinsic, layer-specific sparsity patterns of attention without any fine-tuning.

Robots can think (and act) twice as fast: HeiSD's hybrid speculative decoding turbocharges embodied agents by intelligently switching between draft and retrieval strategies.

Forget treating document graphics as mere pixels: this new OCR system parses them into reusable code, unlocking multimodal supervision and outperforming existing systems.

Stop struggling with compounding errors in long-horizon robotic tasks: AtomVLA leverages LLMs and latent world models to decompose tasks and score actions, boosting success rates to 97% on LIBERO.

VLA models get a 1.73x speedup with only 5-7% overhead thanks to RAPID, a new edge-cloud collaborative inference framework that smartly handles visual noise and motion continuity.

Achieve up to 10.94x speedup in end-to-end latency for on-device agentic RAG by intelligently scheduling tasks across heterogeneous mobile SoC hardware.

By integrating kinematic prediction with speculative decoding, KERV enables VLA models to achieve a 27-37% speedup in robot control tasks without sacrificing success rate.

Attention entropy reveals exploitable sparsity in VAR models, enabling 3.4x faster image generation without sacrificing quality.