张晟源
电子信息博士关注扩散模型与流匹配模型的蒸馏、偏好优化及生成质量评价，研究高效采样、训练稳定性与三维场景补全。
项目与作品
- 人工智能发展简史图谱 内容撰写
研究论文 · 10篇
2026AAAI
Diffusion Distillation With Direct Preference Optimization For Efficient 3D LiDAR Scene Completion
2026IJCV
SCoRE: Standardized Human Evaluation Provides a Reliable Measure for Semantic Consistency of Text-to-Image Generation
2026ECCV
InstaPano: Zero-shot Instance Layout Controlled Panorama Generation Via Global Attention Fusion
2026ICML
Mean Flow Distillation: Robust and Stable Distillation for Flow Matching Models
2026arXiv · 预印本
Beyond Binary Preference: Aligning Diffusion Models to Fine-grained Criteria by Decoupling Attributes
2025ICLR
Distribution Backtracking Builds A Faster Convergence Trajectory for Diffusion Distillation
2025ICCV
Distilling Diffusion Models to Efficient 3D LiDAR Scene Completion
2025ACM MM
Inversion-DPO: Precise and Efficient Post-Training for Diffusion Models
2024AAAI
Reducing Spatial Fitting Error in Distillation of Denoising Diffusion Models
2024arXiv · 预印本