Xuanming Liu — accepted AI-conference papers

2026

Streaming Autoregressive Video Generation via Diagonal Distillation

Jinxiu Liu, Xuanming Liu, Kangfu Mei, Yandong Wen, Ming-Hsuan Yang, Weiyang Liu

ICLR 2026poster

Large-scale pretrained diffusion models have significantly enhanced the quality of generated videos, and yet their use in real-time streaming remains limited. Autoregressive models offer a natural framework for sequential frame synthesis but require heavy computation to achieve high fidelity. Diffus…

Cited by 0 Source ↗code ↗Scholar ↗

2026

XYZFlow: Scaling Multidimensional Shortcut Flows for Efficient Generative Modeling

Jinxiu Liu, Xuanming Liu, Kangfu Mei, Yandong Wen, Weiyang Liu

ICML 2026poster

The pursuit of high-fidelity image generation faces a fundamental trade-off between sampling speed and output quality. While diffusion models excel in quality, their iterative nature incurs high computational costs. Current efficient methods primarily focus on distilling pre-trained models into few-…

Cited by 0 Source ↗Scholar ↗