Ji Ma

Achieving a staggering 98.7% success rate in humanoid loco-manipulation by seamlessly chaining meta-skills through a novel contact flow representation.

Gemini Embedding 2's unified multimodal embeddings beat specialized models across diverse tasks and even generalize zero-shot to niche fields like astronomy and culinary arts.

Correcting a vision-language model's "hallucinations" is now as simple as pinpointing and editing the right intermediate representation, sidestepping costly retraining or dual inference.