SZ
Songchun Zhang
cs.CVcs.GRcs.MMcs.AIcs.LG
On Valency
published · living versionsW_afvan2j4·v1 · currentpublished
WorldStream: Streaming Interactive 3D Worlds with Hierarchical Planning and Persistent State
with Xiaoxiao Ma, Weiyang Jin, Yanwen Ma, Haoran Li +9
1 version
Preprints & journals
20 papers in the corpus · 2023–2026WorldStream: Streaming Interactive 3D Worlds with Hierarchical Planning and Persistent State2610.28827v1 · Xiaoxiao Ma, Weiyang Jin, Yanwen Ma et al.2026 · 0 citationsarXivon Valency
SGF+: Decoupling Gradient Flows for Autoregressive Video Generation2610.10429v3 · Zihan Su, Junhao Zhuang, Yaowei Li et al.2026 · 0 citationsarXiv
Mask Forcing: Improving Autoregressive Video Diffusion Distillation via Dual-Noise Masking Rollout2609.09123v1 · Zhuoran Zhao, Shengju Qian, Tongtong Liang et al.2026 · 0 citationsarXiv
Building Pretraining Data for World Models: An Unreal Engine-Based Pipeline for Action-Conditioned Video Generation2609.03557v1 · Haoyu Wang, Songchun Zhang, Haoran Li et al.2026 · 0 citationsarXiv
EchoWM: Open and Enterable Omnimodal World Models2608.23189v1 · Songchun Zhang, Yaowei Li, Junhao Zhuang et al.2026 · 0 citationsarXiv
Long-Horizon Audio-Visual Generation for Persistent Stories and Interactive Worlds2608.23383v2 · Nan Duan, Haoyang Huang, Weiyang Jin et al.2026 · 0 citationsarXiv
LiveLight: Real-time Streaming Video Relighting with Interactive Control2608.01771v1 · Yue Ma, Jiangming Wang, Yucheng Wang et al.2026 · 0 citationsarXiv
FlexComposer: Unified Video Compositing from Images to Dynamic Footage with Flexible Trajectory Control2607.29627v1 · Songchun Zhang, Sitong Guo, Xianghao Kong et al.2026 · 0 citationsarXiv
Self Gradient Forcing: Native Long Video Extrapolation2607.20368v2 · Junhao Zhuang, Shiyi Zhang, Yuxuan Bian et al.2026 · 0 citationsarXiv
GeoEdit: Geometry-Aware Object Editing via Dual-Branch Denoising2606.30003v1 · Yi He, Jiangming Wang, Xinyu Wang et al.2026 · 0 citationsarXiv
Ultra Flash: Scaling Real-Time Streaming Video Generation to High Resolutions2606.09150v4 · Jie Huang, Xin Lu, Zihao Fan et al.2026 · 0 citationsAdvances in Neural Information Processing Systems 2026
Echo-Memory: A Controlled Study of Memory in Action World Models2606.09803v1 · Wayne King, Zeyue Xue, Yuxuan Bian et al.2026 · 0 citationsarXiv
Echo-Infinity: Learning Evolving Memory for Real-Time Infinite Video Generation2606.04527v1 · Yuxuan Bian, Zeyue Xue, Songchun Zhang et al.2026 · 0 citationsarXiv
Astrolabe: Steering Forward-Process Reinforcement Learning for Distilled Autoregressive Video Models2603.17051v1 · Songchun Zhang, Zeyue Xue, Siming Fu et al.2026 · 0 citationsarXiv
ShotVerse: Advancing Cinematic Camera Control for Text-Driven Multi-Shot Video Creation2603.11421v2 · Songlin Yang, Zhe Wang, Xuyi Yang et al.2026 · 0 citationsarXiv
Composing Concepts from Images and Videos via Concept-prompt Binding2512.09824v2 · Xianghao Kong, Zeyu Zhang, Yuwei Guo et al.2025 · 0 citationsarXiv
SpatialCrafter: Unleashing the Imagination of Video Diffusion Models for Scene Reconstruction from Limited Observations2505.11992v2 · Songchun Zhang, Huiyao Xu, Sitong Guo et al.2025 · 6 citationsarXiv
Pragmatist: Multiview Conditional Diffusion Models for High-Fidelity 3D Reconstruction from Unposed Sparse Views2412.08412v2 · Songchun Zhang, Chunhui Zhao2024 · 1 citationarXiv
Cross-Video Contextual Knowledge Exploration and Exploitation for Ambiguity Reduction in Weakly Supervised Temporal Action Localization2308.12609v2 · Songchun Zhang, Chunhui Zhao2023 · 7 citationsarXiv
Dyna-DepthFormer: Multi-frame Transformer for Self-Supervised Depth Estimation in Dynamic Scenes2301.05871v2 · Songchun Zhang, Chunhui Zhao2023 · 2 citationsarXiv
Career total: 32 works. 20 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.