Scholar
Haoge Deng
Google Scholar ID: S2sbvjgAAAAJ
Institute of Automation, Chinese Academy of Sciences & Beijing Academy of Artificial Intelligence
Citations & Impact
Citations
201
H-index
8
i10-index
6
Publications
9
Co-authors
7
list available
Publications
Resume (English only)
Academic Achievements
- - Emu3.5: A natively multimodal world model that unifies vision and language through end-to-end next-token prediction on interleaved video-derived data, enhanced by reinforcement learning and DiDA-based parallel decoding for efficient, spatiotemporally consistent generation.
- - Uniform Discrete Diffusion with Metric Path for Video Generation: A simple yet powerful discrete framework that formulates video generation as an iterative process of global refinement over spatiotemporal tokens, enabling efficient scaling to long-duration videos.
- - Autoregressive Video Generation without Vector Quantization: A non-quantized autoregressive model that enables efficient video generation by reformulating the video creation as frame-by-frame and set-by-set predictions.
- - You See it, You Got it: Learning 3D Creation on Pose-Free Videos at Scale: A scalable visual-conditional MVD model for open-world 3D creation, which can be trained on web-scale video collections without camera pose annotations.
- - GeoDream: Disentangling 2D and Geometric Priors for High-Fidelity and Consistent 3D Generation: A 3D generation method that integrates explicit generalized 3D priors with 2D diffusion priors to enhance the capability of obtaining unambiguous 3D consistent geometric structures without sacrificing diversity or fidelity.
- - SketchKnitter: Vectorized Sketch Generation with Diffusion Models: A method that achieves vectorized sketch generation by reversing the stroke deformation process using a diffusion model learned from real sketches.
Research Experience
- Current research work during his PhD focuses on generative models and multimodal generation.
Education
- - PhD student, jointly supervised by the Institute of Automation, Chinese Academy of Sciences (CASIA), and Beijing Academy of Artificial Intelligence (BAAI), supervised by Prof. Zhaoxiang Zhang and Dr. Xinlong Wang
- - MSc degree, BUPT, supervised by Prof. Yonggang Qi
- - Bachelor's degree in Electronics Information Science and Technology, BUPT, 2022
Background
- Research interests include generative models, with a particular focus on multimodal generation.
Miscellany
- Contact: Email / Scholar / Github
