8 papers
Composed Object Retrieval: Object-level Retrieval via Composed Expressions
Tong Wang, Guanyu Yang, Nian Liu +4
Retrieving fine-grained visual content based on user intent remains a challenge in multimodal systems. Although current Composed Image Retrieval (CIR) methods combine reference ima…
UAOR: Uncertainty-aware Observation Reinjection for Vision-Language-Action Models
Jiabing Yang, Yixiang Chen, Yuan Xu +14
Vision-Language-Action (VLA) models leverage pretrained Vision-Language Models (VLMs) as backbones to map images and instructions to actions, demonstrating remarkable potential for…
AURORA:Augmented Understanding via Structured Reasoning and Reinforcement Learning for Reference Audio-Visual Segmentation
Ziyang Luo, Nian Liu, Fahad Shahbaz Khan +1
Reference Audio-Visual Segmentation (Ref-AVS) tasks challenge models to precisely locate sounding objects by integrating visual, auditory, and textual cues. Existing methods often…
TAViS: Text-bridged Audio-Visual Segmentation with Foundation Models
Ziyang Luo, Nian Liu, Xuguang Yang +5
Audio-Visual Segmentation (AVS) faces a fundamental challenge of effectively aligning audio and visual modalities. While recent approaches leverage foundation models to address dat…
Diversity Has Always Been There in Your Visual Autoregressive Models
Tong Wang, Guanyu Yang, Nian Liu +6
Visual Autoregressive (VAR) models have recently garnered significant attention for their innovative next-scale prediction paradigm, offering notable advantages in both inference e…
RAGNet: Large-scale Reasoning-based Affordance Segmentation Benchmark towards General Grasping
Dongming Wu, Yanping Fu, Saike Huang +8
General robotic grasping systems require accurate object affordance perception in diverse open-world scenarios following human instructions. However, current studies suffer from th…