Shi-Min Hu

RoboDawn is investigated, a human-intuitive interface that exposes robotic control to an agentic VLM through a compact set of discrete translation, rotation, and gripper commands, and an in-context learning (ICL) scheme is introduced that ground the VLM in both interface usage and task-solving strategies.

GUICrafter achieves superior GUI agent performance with just a fraction of the data, revolutionizing the way we think about training in data-scarce environments.

TopoCap enables seamless motion retargeting across arbitrary skeletal structures, outperforming traditional methods and opening the door to unprecedented animation flexibility.

LLMs can now optimize CUDA kernels across diverse scientific computing and LLM workloads, rivaling hand-tuned libraries like cuBLAS.

Forget iterative optimization: now you can edit 3D models in a single feedforward pass with globally consistent deformations and high-fidelity textures.

Achieve zero package hallucinations from LLMs in dependency recommendation by monitoring the decoding process and intervening with an authoritative package list.

High-quality data is all it takes: Bee-8B, trained on the new Honey-Data-15M dataset, leapfrogs existing fully open MLLMs to rival semi-open models.