2 papers
cs.CV2026
TraceCLIP: Recovering Local Semantics from Patch-to-CLS Contributions
Xinran Liu, Shouqian Shi, Yutong Chen +3
Dense vision-language understanding, including object localization, region recognition, and open-vocabulary semantic segmentation, requires associating language concepts with spati…
cs.CL2026
IRIS: Reusable Identity Representations from Frozen LLMs for Entity Alignment
Xinran Liu, Shengtao Li, Shouqian Shi +2
Entity alignment (EA) identifies entities across knowledge graphs (KGs) that refer to the same real-world object. Conventional EA methods mainly exploit explicit graph structures a…