2 papers
cs.CV2026
Reading Images Like Texts: Sequential Image Understanding in Vision-Language Models
Yueyan Li, Chenggong Zhao, Zeyuan Zang +2
Vision-Language Models (VLMs) have demonstrated remarkable performance across a variety of real-world tasks. However, existing VLMs typically process visual information by serializ…
cs.LG2025
Fine-Tuning is Subgraph Search: A New Lens on Learning Dynamics
Yueyan Li, Wenhao Gao, Caixia Yuan +1
The study of mechanistic interpretability aims to reverse-engineer a model to explain its behaviors. While recent studies have focused on the static mechanism of a certain behavior…