Papers

Work across generative modeling, scientific machine learning, and learning systems.

When Local Rules Create Global Order: Self-Organized Representation Learning for Latent Diffusion Models [pdf] [supp]

Junrong Lian, Weijian Deng, Pengxu Wei, Yaqin Chen, Qixiang Ye, Liang Lin

CVPR 2026

Training-Free Guided Generation techniques for Conditional Flow Matching on Scientific Datasets [arxiv] [pdf]

Meet Hemant Parikh, Yaqin Chen, Jianxun Wang

Benchmarking In-context Experiential Learning Through Repeated Product Recommendations [arxiv] [pdf]

Gilbert Yang*, Yaqin Chen*, Thomson Yen*, Hongseok Namkoong

Submitted to NeurIPS 2026

MVU-AE: Minimum Variance Unbiased Autoencoder [pdf]

Yaqin Chen, Junrong Lian, Qing Zhao, Xiangyang Ji, Liang Lin, Pengxu Wei