Li jiaye — accepted AI-conference papers

2026

Pyramid Patchification Flow for Visual Generation

Hui Li, Baoyou Chen, Li jiaye, Jingdong Wang, Siyu Zhu

ICLR 2026poster

Diffusion Transformers (DiTs) typically use the same patch size for $\operatorname{Patchify}$ across timesteps, enforcing a constant token budget across timesteps. In this paper, we introduce Pyramidal Patchification Flow (PPFlow), which reduces the number of tokens for high-noise timesteps to impro…

Cited by 0 Source ↗code ↗Scholar ↗

2025

Retaining Knowledge and Enhancing Long-Text Representations in CLIP through Dual-Teacher Distillation

Yuheng Feng, Changsong Wen, Zelin Peng, Li jiaye, Siyu Zhu

CVPR 2025poster

Contrastive language-image pretraining models such as CLIP have demonstrated remarkable performance in various text-image alignment tasks. However, the inherent 77-token input limitation and reliance on predominantly short-text training data restrict its ability to handle long-text tasks effectively…

Cited by 0 Source ↗PDF ↗Scholar ↗