Xiangyu Yue

发表论文 6 篇 · 总被引 105 次 · h-index 4

代表论文

• DiTCtrl: Exploring Attention Control in Multi-Modal Diffusion Transformer for Tuning-Free Multi-Prompt Longer Video Generation (2024 · Computer Vision and Pattern Recognition · 被引 80)
• Customize Your Visual Autoregressive Recipe with Set Autoregressive Modeling (2024 · arXiv.org · 被引 20)
• Learning to Integrate Diffusion ODEs by Averaging the Derivatives (2025 · Neural Information Processing Systems · 被引 8)
• Training Matting Models without Alpha Labels (2024 · AAAI Conference on Artificial Intelligence · 被引 6)
• VideoCanvas: Unified Video Completion from Arbitrary Spatiotemporal Patches via In-Context Conditioning (2025 · arXiv.org · 被引 4)
• Amortized Moment Matching for Visual Generation (2026)