6 papers
SSQA: Sibling-Selective Quadtree Attention for Hierarchical Modeling in Perception Tasks
Yufan Chen, Arnav Bali, Angela Liu +2
Perception tasks for navigation in robotics, including aerial platforms such as drones and autonomous driving systems, are inherently structured. Drone-mounted cameras typically ca…
Cognitive Mismatch in Multimodal Large Language Models for Discrete Symbol Understanding
Yinghui Li, Jiayi Kuang, Peng Xing +11
Multimodal large language models (MLLMs) perform strongly on natural images, yet their ability to understand discrete visual symbols remains unclear. We present a multi-domain benc…
TangramPuzzle: Evaluating Multimodal Large Language Models with Compositional Spatial Reasoning
Daixian Liu, Jiayi Kuang, Yinghui Li +8
Multimodal Large Language Models (MLLMs) have achieved remarkable progress in visual recognition and semantic understanding, yet precise compositional spatial reasoning under geome…
Open-World Test-Time Training: Self-Training with Contrast Learning
Houcheng Su, Mengzhu Wang, Jiao Li +3
Traditional test-time training (TTT) methods, while addressing domain shifts, often assume a consistent class set, limiting their applicability in real-world scenarios characterize…
Singular Value Penalization and Semantic Data Augmentation for Fully Test-Time Adaptation
Houcheng Su, Daixian Liu, Mengzhu Wang +1
Fully test-time adaptation (FTTA) adapts a model that is trained on a source domain to a target domain during the testing phase, where the two domains follow different distribution…
Graph based Label Enhancement for Multi-instance Multi-label learning
Houcheng Su, Jintao Huang, Daixian Liu +3
Multi-instance multi-label (MIML) learning is widely applicated in numerous domains, such as the image classification where one image contains multiple instances correlated with mu…