Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Pengfei Wan

Pengfei Wan

3 indexed papers

Recent (6 mo)
3
With code
0
Influential cites
0
Benchmarked
0

Publications per year

3
26

Top categories

Vision×2AI×1ML×1

Frequent co-authors

Junhao Cheng1×
Liang Hou1×
Tianxiong Zhong1×
Xin Tao1×
Kun Gai1×
Jing Liao1×

Research Timeline

2026
Regularized Offline Policy Optimization with Posterior Hybrid Bayesian Belief

The paper introduces Posterior Hybrid Bayesian Belief (PhyB), a novel framework that reformulates policy optimization in Bayesian Offline RL by approximating expectations as a convex combination over a subset of dynamics models, achieving state-of-the-art performance.

VLMs are Good Teachers for Video Reasoning via Adaptive Test-Time Optimization

The paper proposes using Vision-Language Models (VLMs) as 'teachers' to guide Video Generation Models (VGMs) during test-time optimization, significantly improving video reasoning capabilities.

Geometry-Aware Implicit Memory for Video World Models

The paper proposes GIM-World, a geometry-aware implicit memory framework that significantly improves long-horizon video world models by explicitly encoding 3D scene geometry into a compact memory state.

Highlighted terms show continued research focus across papers

Papers

cs.CVRecentJun 1, 2026

VLMs are Good Teachers for Video Reasoning via Adaptive Test-Time Optimization

Junhao Cheng, Liang Hou, Tianxiong Zhong, Xin Tao +3 more

The paper proposes using Vision-Language Models (VLMs) as 'teachers' to guide Video Generation Models (VGMs) during test-time optimization, significantly improving video reasoning capabilities.

View →
cs.CVRecentJun 1, 2026

Geometry-Aware Implicit Memory for Video World Models

Zhengxuan Wei, Xu Guo, Xinghui Li, Xunzhi Xiang +7 more

The paper proposes GIM-World, a geometry-aware implicit memory framework that significantly improves long-horizon video world models by explicitly encoding 3D scene geometry into a compact memory stat…

View →
cs.AIcs.LGRecentMay 30, 2026

Regularized Offline Policy Optimization with Posterior Hybrid Bayesian Belief

Hongqiang Lin, Pengfei Wang, Nenggan Zheng

The paper introduces Posterior Hybrid Bayesian Belief (PhyB), a novel framework that reformulates policy optimization in Bayesian Offline RL by approximating expectations as a convex combination over…

View →