papers

Publications (11)

cs.RO2025

CushionCatch: A Compliant Catching Mechanism for Mobile Manipulators via Combined Optimization and Learning

Bingjie Chen, Keyu Fan, Qi Yang +6

Catching flying objects with a cushioning process is a skill commonly performed by humans, yet it remains a significant challenge for robots. In this paper, we present a framework…

cs.CV2025

MonoSplat: Generalizable 3D Gaussian Splatting from Monocular Depth Foundation Models

Yifan Liu, Keyu Fan, Weihao Yu +3

Recent advances in generalizable 3D Gaussian Splatting have demonstrated promising results in real-time high-fidelity rendering without per-scene optimization, yet existing approac…

cs.CV2025

GEN3D: Generating Domain-Free 3D Scenes from a Single Image

Yuxin Zhang, Ziyu Lu, Hongbo Duan +5

Despite recent advancements in neural 3D reconstruction, the dependence on dense multi-view captures restricts their broader applicability. Additionally, 3D scene generation is vit…

cs.LG2026

Attention Sink in Transformers: A Survey on Utilization, Interpretation, and Mitigation

Zunhai Su, Hengyuan Zhang, Wei Wu +24

As the foundational architecture of modern machine learning, Transformers have driven remarkable progress across diverse AI domains. Despite their transformative impact, a persiste…

cs.RO2025

DyPho-SLAM : Real-time Photorealistic SLAM in Dynamic Environments

Yi Liu, Keyu Fan, Bin Lan +1

Visual SLAM algorithms have been enhanced through the exploration of Gaussian Splatting representations, particularly in generating high-fidelity dense maps. While existing methods…

cs.CV2026

Principled RL for Flow Matching Emerges from the Chunk-level Policy Optimization

Yifu Luo, Haoyuan Sun, Xinhao Hu +12

Recent Progress in post-training flow matching for text-to-image (T2I) generation with Group Relative Policy Optimization (GRPO) has demonstrated strong potential. However, it is h…

cs.CV2026

Reinforcement Learning Meets Masked Generative Models: Mask-GRPO for Text-to-Image Generation

Yifu Luo, Xinhao Hu, Keyu Fan +6

Reinforcement learning (RL) has garnered increasing attention in text-to-image (T2I) generation. However, most existing RL approaches are tailored to either diffusion models or aut…

cs.CV2026

XStreamVGGT: Extremely Memory-Efficient Streaming Vision Geometry Grounded Transformer with KV Cache Compression

Zunhai Su, Weihao Ye, Hansen Feng +5

Learning-based 3D visual geometry models have benefited substantially from large-scale transformers. Among these, StreamVGGT leverages frame-wise causal attention for strong stream…

cs.CV2026

XStreamVGGT: Extremely Memory-Efficient Streaming Vision Geometry Grounded Transformer with KV Cache Compression

Zunhai Su, Weihao Ye, Hansen Feng +5

Learning-based 3D visual geometry models have significantly advanced with the advent of large-scale transformers. Among these, StreamVGGT leverages frame-wise causal attention to d…

cs.RO2026

GGD-SLAM: Monocular 3DGS SLAM Powered by Generalizable Motion Model for Dynamic Environments

Yi Liu, Haoxuan Xu, Hongbo Duan +5

Visual SLAM algorithms achieve significant improvements through the exploration of 3D Gaussian Splatting (3DGS) representations, particularly in generating high-fidelity dense maps…

cs.CV2025

Hear-Your-Click: Interactive Object-Specific Video-to-Audio Generation

Yingshan Liang, Keyu Fan, Zhicheng Du +5

Video-to-audio (V2A) generation shows great potential in fields such as film production. Despite significant advances, current V2A methods relying on global video information strug…