Junliang Ye

PolyFlow: Continuous Topology Embedding Flow Matching for Artist-style Mesh Generation

Chunshi Wang*, Haohan Weng*, Junliang Ye*, Biwen Lei, Yang Li, Zibo Zhao, et al.

arXiv preprint, 2026

A Transformer-based flow-matching framework for parallel artist-style mesh generation with continuous topology embedding, achieving faster inference and precise vertex-count control.

PhysForge: Generating Physics-Grounded 3D Assets for Interactive Virtual World

Yunhan Yang*, Chunshi Wang*, Junliang Ye*, Yang Li, Zanxin Chen, Zehuan Huang, et al.

International Conference on Machine Learning (ICML), 2026

A decoupled two-stage framework supported by PhysDB, a large-scale dataset of 150,000 assets with four-tier physical annotations.

Code Project

UniVerse3D: Emerging Properties of Unified Multimodal Models in 3D Understanding and Generation

Junliang Ye*, Zehuan Huang*, Yansong Qu*, Chunshi Wang, Yunhan Yang, Yang Li, et al.

CVPR FINDINGS Track, 2026

We propose UniVerse3D, the first 3D Unified multimodal models.

Nano3D: A Training-Free Approach for Efficient 3D Editing Without Masks

Junliang Ye*, Shenghao Xie*, Ruowen Zhao, Zhengyi Wang, Hongyu Yan, Wenqiang Zu, et al.

International Conference on Learning Representations (ICLR), 2026

A training-free framework for precise and coherent 3D object editing without masks.

Paper Code Project

Part-X-MLLM: Part-aware 3D Multimodal Large Language Model

Chunshi Wang*, Junliang Ye†*, Yunhan Yang*, Yang Li, Zizhuo Lin, Jun Zhu, et al.

International Conference on Learning Representations (ICLR), 2026

A native 3D multimodal large language model that unifies diverse 3D tasks via structured, executable grammar.

Paper Code Project

DeepMesh: Auto-Regressive Artist-Mesh Creation With Reinforcement Learning

Ruowen Zhao*, Junliang Ye*, Zhengyi Wang*, Guangce Liu, Yiwen Chen, Yikai Wang, et al.

IEEE International Conference on Computer Vision (ICCV), 2025

Generates meshes with intricate details and precise topology, surpassing SOTA in both precision and quality.

Paper Code Project

DreamReward: Aligning Human Preference in Text-to-3D Generation

Junliang Ye*, Fangfu Liu*, Qixiu Li, Zhengyi Wang, Yikai Wang, Xinzhou Wang, et al.

European Conference on Computer Vision (ECCV), 2024

A comprehensive framework to learn and improve text-to-3D models from human preference feedback.

Paper Code Project

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

DreamReward-X: Boosting High-Quality 3D Generation with Human Preference Alignment

Fangfu Liu, Junliang Ye, Hanyang Wang, Zhengyi Wang, Jun Zhu, Yueqi Duan

IEEE TPAMI, 2025

AnimatableDreamer: Text-Guided Non-rigid 3D Model Generation and Reconstruction

Xinzhou Wang, Yikai Wang, Junliang Ye, Fucun Sun, Zhengyi Wang, Ling Wang, et al.

European Conference on Computer Vision (ECCV), 2024

GEM: Generative Supervision Helps Embodied Intelligence

Ruowen Zhao, Bangguo Li, Zuyan Liu, Yinan Liang, Junliang Ye, et al.

European Conference on Computer Vision (ECCV), 2026

DeepMesh-v2: Auto-Regressive Artist-Mesh Creation With Reinforcement Learning

Junliang Ye*, Ruowen Zhao*, Zhengyi Wang*, Yikai Wang, Jun Zhu

arXiv preprint, 2025