Junliang Ye

Junliang Ye | 叶俊良

I am a third-year master's student in the Department of Computer Science at Tsinghua University , advised by Prof. Jun Zhu. In 2022, I obtained my B.S. in the School of Mathematical Sciences at Peking University.

My research interests lie in computer vision (e.g., 3D AIGC and video generation), multimodal large models (e.g., native large models), and reinforcement learning from human feedback (DPO, GRPO).

My email:yejl23@mails.tsinghua.edu.cn

Email / CV / Google Scholar / Github

News

2026-02: One papers accepted by TIP 2026.

2026-01: Two papers on 3D-MLLM are accepted by ICLR 2026.

2025-09: One papers on 3D-MLLM are accepted by NeurIPS 2025.

2025-09: One papers on 3D Vision are accepted by TPAMI 2025.

2025-06: One papers on 3D Vision are accepted by ICCV 2025.

2024-07: Two papers on 3D AIGC are accepted by ECCV 2024.

Preprints

*equal contribution †Project leader

DeepMesh-v2: Auto-Regressive Artist-Mesh Creation With Reinforcement Learning
Junliang Ye*, Ruowen Zhao*, Zhengyi Wang*, et. al
Arxiv, 2025
[arXiv] [Code] [Project Page]

We propose DeepMesh-v2, which generates meshes with intricate details and precise topology, surpassing state-of-the-art methods in both precision and quality.

Selected Research

	NANO3D: A Training-Free Approach for Efficient 3D Editing Without Masks Junliang Ye, Shenghao Xie, Ruowen Zhao, Zhengyi Wang, Hongyu Yan, et. al International Conference on Learning Representations (ICLR), 2026 [arXiv] [Code] [Project Page] we propose Nano3D, a training-free framework for precise and coherent 3D object editing without masks.
	Part-X-MLLM: Part-aware 3D Multimodal Large Language Model Chunshi Wang, Junliang Ye†, Yunhan Yang, et. al International Conference on Learning Representations* (ICLR), 2026 [arXiv] [Code] [Project Page] We introduce Part-X-MLLM, a native 3D multimodal large language model that unifies diverse 3D tasks by formulating them as programs in a structured, executable grammar.
	ShapeLLM-Omni: A Native Multimodal LLM for 3D Generation and Understanding Junliang Ye, Zhengyi Wang, Ruowen Zhao, Shenghao Xie, Jun Zhu Conference on Neural Information Processing Systems* (NeurIPS), 2025 (Spotlight) [arXiv] [Code] [Project Page] We propose ShapeLLM-Omni, a multimodal large model that integrates 3D generation, understanding, and editing capabilities.
	DeepMesh: Auto-Regressive Artist-Mesh Creation With Reinforcement Learning Ruowen Zhao, Junliang Ye, Zhengyi Wang, et. al IEEE International Conference on Computer Vision* (ICCV), 2025 [arXiv] [Code] [Project Page] We propose DeepMesh, which generates meshes with intricate details and precise topology, surpassing state-of-the-art methods in both precision and quality.
	DreamReward: Aligning Human Preference in Text-to-3D Generation Junliang Ye, Fangfu Liu, Qixiu Li, Zhengyi Wang, et. al European Conference on Computer Vision (ECCV), 2024 [arXiv] [Code] [Project Page] We present a comprehensive framework, coined DreamReward, to learn and improve text-to-3D models from human preference feedback.

Additional Projects

	ReconX: Reconstruct Any Scene from Sparse Views with Video Diffusion Model Fangfu Liu, Wenqiang Sun, Hanyang Wang, Yikai Wang, Haowen Sun, Junliang Ye, et. al IEEE Transactions on Image Processing* (TIP), 2026 [arXiv] [Code] [Project Page] In this paper, we propose ReconX, a novel 3D scene reconstruction paradigm that reframes the ambiguous reconstruction challenge as a temporal generation task. The key insight is to unleash the strong generative prior of large pre-trained video diffusion models for sparse-view reconstruction.
	DreamReward-X: Boosting High-Quality 3D Generation with Human Preference Alignment Fangfu Liu, Junliang Ye, et. al IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025 [arXiv] [Code] [Project Page] We present a comprehensive framework, coined DreamReward++, where we introduce a reward-aware noise sampling strategy to unleash text-driven diversity during the generation process while ensuring human preference alignment. Grounded by theoretical proof and extensive experiment comparisons, our method successfully generates high-fidelity and diverse 3D results with significant boosts in prompt alignment with human intention. Our results demonstrate the great potential for learning from human feedback to improve 3D generation.
	AnimatableDreamer: Text-Guided Non-rigid 3D Model Generation and Reconstruction with Canonical Score Distillation Xinzhou Wang, Yikai Wang, Junliang Ye, et. al European Conference on Computer Vision (ECCV), 2024 [arXiv] [Code] [Project Page] We propose ANIMATABLEDREAMER, a framework with the capability to generate generic categories of non-rigid 3D models.
	PKU_WICT at TRECVID 2022: Disaster Scene Description and Indexing Task Yanzhe Chen, HsiaoYuan Hsu, Junliang Ye, Zhiwen Yang, Zishuo Wang, Xiangteng He, Yuxin Peng Virtual, Online [arXiv] [Code] [Project Page] We achieved first place in the TRECVID 2022 competition.

Academic Services

Review for CVPR 2023, ICLR 2026.

Website Template