4 papers
ReVSI: Rebuilding Visual Spatial Intelligence Evaluation for Accurate Assessment of VLM 3D Reasoning
Yiming Zhang, Jiacheng Chen, Jiaqi Tan +3
Current evaluations of spatial intelligence can be systematically invalid under modern vision-language model (VLM) settings. First, many benchmarks derive question-answer (QA) pair…
Instruction Guided Multi Object Image Editing with Quantity and Layout Consistency
Jiaqi Tan, Fangyu Li, Yang Liu
Instruction driven image editing with standard CLIP text encoders often fails in complex scenes with many objects. We present QL-Adapter, a framework for multiple object editing th…
Robust Multimodal Semantic Segmentation with Balanced Modality Contributions
Jiaqi Tan, Xu Zheng, Fangyu Li +1
Multimodal semantic segmentation enhances model robustness by exploiting cross-modal complementarities. However, existing methods often suffer from imbalanced modal dependencies, w…
RMMSS: Towards Advanced Robust Multi-Modal Semantic Segmentation with Hybrid Prototype Distillation and Feature Selection
Jiaqi Tan, Xu Zheng, Yang Liu
Multi-modal semantic segmentation (MMSS) faces significant challenges in real-world applications due to incomplete, degraded, or missing sensor data. While current MMSS methods typ…