Scholar

Xiangyu Yue

Google Scholar ID: -xQ-C1sAAAAJ

The Chinese University of Hong Kong / UC Berkeley / Stanford University / NJU

Artificial IntelligenceComputer VisionMulti-modal Learning

Homepage↗Google Scholar↗

Citations & Impact

All-time

Citations

11,719

H-index

33

i10-index

57

Publications

20

Co-authors

31

list available

Publications

95 items

Position Forcing: Self-Conditioning 3D Generation

2026

Cited

0

MC-Sparse: Deconstructing and Closing the Dense-Sparse Attention Gap in Diffusion Transformers

2026

Cited

0

Vela: Scaling Vision-Language-Action Models with Adaptive Action Curve Parametrization

2026

Cited

0

EviRover: Reinforcing Agentic Perception Beyond a Glance

2026

Cited

0

Does Native 3D Texture Generation Necessarily Require 3D Assets for Training?

2026

Cited

0

Watch, Recall, Act: Always-On Robots in Concurrent Embodied Streams

2026

Cited

0

Evolve Vision-Language-Action Model into an Agent with On-the-fly Tool-use

2026

Cited

0

From Local Mismatch to Global Impact: Optimizing Cache Reuse Policy for Efficient Diffusion

2026

Cited

0

Co-authors

17 total

Professor of the Graduate School, EECS, University of California, Berkeley

Alberto Sangiovanni Vincentelli

Professor of EECS, University of California, Berkeley

Sanjit A. Seshia

Professor of EECS, University of California, Berkeley

Wanli Ouyang (欧阳万里)

Shanghai AI Lab & CUHK

Professor of Shanghai AI Laboratory; Shenzhen Institutes of Advanced Technology, CAS

Boston University, Google

Hongsheng Li (李鸿升)

The Chinese University of Hong Kong

Ph.D. of National University of Singapore