Hengyuan Zhang

Generating geometrically consistent digital twins from video just got easier: Pantheon360 leverages 360° video diffusion and a 3D cache to achieve superior results.

Attention Sink, where Transformers fixate on seemingly irrelevant tokens, is more than just a quirk – it's a fundamental challenge impacting training, inference, and even causing hallucinations, demanding a systematic approach to understanding and mitigating its effects.

Achieve better compression in low-bit quantization by considering not just numerical sensitivity, but also the structural role of each layer.