Zichen Wen

Advanced
@ZichenWen1

Ph.D student at SJTU

scholar.google.com/citations?user=N-aPFvEAAAAJ&hl=zh-CN

DART. [EMNLP 2025 main 🔥] Code for "Stop Looking for Important Tokens in Multimodal Language Models: Duplication Matters More"

121

DIJA. (ICLR 2026 🔥) Code for "The Devil behind the mask: An emergent safety vulnerability of Diffusion LLMs"

79

EPIC. (NeurIPS 2025 🔥) Official implementation for "Efficient Multi-modal Large Language Models via Progressive Consistency Distillation"

49

AHGFC. The source code for “Homophily-Related: Adaptive Hybrid Graph Filter for Multi-View Graph Clustering”

11
4
Apply