孟辰烨
电子信息博士2024年起在浙江大学攻读博士学位。
2019—2023年就读于江南大学数字媒体技术专业，曾连续两年获国家奖学金。
关注生成模型的评价与偏好对齐，研究场景图、细粒度属性和高效蒸馏如何改善生成内容。
项目与作品
- 人工智能发展简史图谱 内容撰写
- PoemPalette 团队成员
研究论文 · 10篇
2026CHI
PoemPalette: Facilitating Poetry Creative Exploration and Foundational Understanding through the Ideorealm Alignment of Paintings and Poems
2026ACM MM
SEPO: Self-Evolving Preference Optimization for Diffusion Alignment via Dynamic Rule Refinement
2026CVPR
Circular-DPO: Aligning Multi-Stage 3D Generative Models via Preference Feedback Loop
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
2025FITEE
Image Generation Evaluation: A Comprehensive Survey of Human and Automatic Evaluations
2024arXiv · 预印本
LAION-SG: An Enhanced Large-Scale Dataset for Training Complex Image-Text Models with Structural Annotations
2023IJCAI
Learning Object Consistency and Interaction in Image Generation from Scene Graphs
学术荣誉
江南大学2023届至善荣誉生“特别荣誉奖”