Bojun Yang — accepted AI-conference papers

2026

SARA: Semantic-Anchored Referential Alignment for Ego–Exo Instance Correspondence

Yueyang Ge, Ye Lin, Bojun Yang, Yilin Huang, Yi Guo

IJCAI 2026

Achieving visual coordination between egocentric (ego) and exocentric (exo) perspectives is a cornerstone of augmented reality and human-machine collaboration. However, extreme viewpoint shifts and occlusions often cause semantic drift in existing methods, making robust cross-view correspondence a f

Cited by 0 Scholar ↗

2026

Test-Time Attention Purification for Backdoored Large Vision Language Models

Zhifang Zhang, Bojun Yang, Shuo He, Weitong Chen, Wei Emma Zhang, Olaf Maennel +2

CVPR 2026

Despite the strong multimodal performance, large vision-language models (LVLMs) are vulnerable during fine-tuning to backdoor attacks, where adversaries insert trigger-embedded samples into the training data to implant behaviors that can be maliciously activated at test time. Existing defenses typic

Cited by 0 Source ↗Scholar ↗