Jinghao Guo

Multi-teacher distillation fails not because of bad routing, but because high-variance instruction feedback quietly drowns out precision domains like mathematics during shared student updates.

Mage-VL slashes visual token usage by over 75% while enhancing real-time multimodal performance, outperforming larger models in video understanding.

Mage-Flow achieves high-resolution image generation and editing in under a second on a single GPU, challenging the notion that larger models are always necessary for quality.