3 papers
cs.LG2025
GNSP: Gradient Null Space Projection for Preserving Cross-Modal Alignment in VLMs Continual Learning
Tiantian Peng, Yuyang Liu, Shuo Yang +2
Contrastive Language-Image Pretraining has demonstrated remarkable zero-shot generalization by aligning visual and textual modalities in a shared embedding space. However, when con…
cs.CV2025
Look-Back: Implicit Visual Re-focusing in MLLM Reasoning
Shuo Yang, Yuwei Niu, Yuyang Liu +3
Multimodal Large Language Models (MLLMs) have achieved remarkable progress in multimodal reasoning. However, they often excessively rely on textual information during the later sta…
cs.CV2024
CASA: Class-Agnostic Shared Attributes in Vision-Language Models for Efficient Incremental Object Detection
Mingyi Guo, Yuyang Liu, Zhiyuan Yan +3
Incremental object detection is fundamentally challenged by catastrophic forgetting. A major factor contributing to this issue is background shift, where background categories in s…