MZ

Mingzhe Zheng

cs.CVcs.AI

On Valency

published · living versions

W_gg6afuje·v1 · currentpublished

Do Agentic Policies Need to Look at Every Step? Geometric Abstraction for Reasoning and Action

with Yihang Liu, Yiming Chen

1 version

Preprints & journals

7 papers in the corpus · 2026

An Empirical Study of Training Pixel-Space Text-to-Image Diffusion Models2608.16887v1 · Dengyang Jiang, Ruoyi Du, Zhennan Chen et al.2026 · 0 citationsarXiv

Twins: Learn to Predict Unified Representations with Focal Loss2607.22531v1 · Kaixiong Gong, Xin Cai, Bin Lin et al.2026 · 0 citationsarXiv

D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models2605.05204v3 · Dengyang Jiang, Xin Jin, Dongyang Liu et al.2026 · 0 citationsarXiv

Group Editing: Edit Multiple Images in One Go2603.22883v3 · Yue Ma, Xinyu Wang, Qianli Ma et al.2026 · 0 citationsarXiv

Manifold-Aware Exploration for Reinforcement Learning in Video Generation2603.21872v1 · Mingzhe Zheng, Weijie Kong, Yue Wu et al.2026 · 0 citationsarXiv

FastVMT: Eliminating Redundancy in Video Motion Transfer2602.05551v3 · Yue Ma, Zhikai Wang, Tianhao Ren et al.2026 · 0 citationsarXiv

iFSQ: Improving FSQ for Image Generation with 1 Line of Code2601.17124v2 · Bin Lin, Zongjian Li, Yuwei Niu et al.2026 · 0 citationsarXiv

Career total: 7 works. 7 are in this corpus.

Profile built from the corpus for this byline.

Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.