Guangrun Wang

Chief Scientist at X-Era AI. Associate Professor and Young Scientist at Sun Yat-sen University. Former Research Fellow at the University of Oxford.

[Guangrun Wang @wanggrun]

@fly51fly This paper has been accepted to CVPR 2026: Weiqi Li, Quande Zhang, Ruifeng Zhai, Liang Lin, Guangrun Wang, VLA Models Are More Generalizable Than You Think: Revisiting Physical and Spatial Modeling

[Guangrun Wang @wanggrun]

@fly51fly This paper has been accepted to CVPR 2026: Weiqi Li, Quande Zhang, Ruifeng Zhai, Liang Lin, Guangrun Wang, VLA Models Are More Generalizable Than You Think: Revisiting Physical and Spatial Modeling

[Guangrun Wang @wanggrun]

@OWW RADAR: Benchmarking Vision-Language-Action Generalization via Real-World Dynamics, Spatial-Physical Intelligence, and Autonomous Evaluation

[Robotics Papers @OWW]

RADAR: Benchmarking Vision-Language-Action Generalization via Real-World Dynamics, Spatial-Physical Intelligence, and Autonomous Evaluation

Yuhao Chen, Zhihao Zhan, Xiaoxin Lin, Zijian Song, Hao Liu, Qinhan Lyu, Yubo Zu, Xiao Chen, …
https://t.co/N7PBjCsf2H [𝚌𝚜.𝚁𝙾]

[Guangrun Wang @wanggrun]

To address this, we propose a one-shot adaptation framework that recalibrates visual representations through lightweight learnable updates. We apply a global affine transformation to visual tokens and improves Libero viewpoint accuracy from 48.5% to 87.1% with only 4K parameters.

[Guangrun Wang @wanggrun]

VLA Models Are More Generalizable Than You Think: Revisiting Physical and Spatial Modeling.

https://t.co/LxFPO43AlN

[Guangrun Wang @wanggrun]

Vision-language-action (VLA) models achieve strong in-distribution performance but degrade sharply under novel camera viewpoints and visual perturbations. We show that this brittleness primarily arises from misalignment in Spatial Modeling, rather than Physical Modeling.

[Guangrun Wang @wanggrun]

For the first time, this mechanism faithfully transplants the principles of diffusion into the world of discrete symbols, achieving true “symbol-level diffusion.”

[Guangrun Wang @wanggrun]

https://t.co/Dsz3Ffliwl

Authentic Discrete Diffusion Model

[Guangrun Wang @wanggrun]

2.Introducing time-weighted cross-entropy loss during training to avoid the over-smoothing problem caused by MSE;
3. Using an argmax + one-hot feedback iteration during inference to achieve true step-by-step discrete denoising.

[Guangrun Wang @wanggrun]

Are diffusion language models all wrong? SYSU team proposes the first authentic discrete diffusion model

[Guangrun Wang @wanggrun]

https://t.co/Dsz3Ffliwl

Authentic Discrete Diffusion Model