📰 News

• [2026-07] 🔥 We release "LingBot-Video", a large-scale MoE video foundation model for embodied intelligence with post-training reward and RL alignment.
• [2026-02] Two papers are accepted by CVPR 2026, Style-GRPO and SpatialT2I.
• [2025-11] Two papers are accepted by AAAI 2026, "NeuralGS" and "Next Patch Prediction".
• [2025-09] A paper is accepted by NeurIPS 2025, "GS2E".
• [2025-07] A paper is accepted by ACM MM 2025, "E-4DGS".
• [2025-06] Two papers are accepted by ICCV 2025, "EvaGaussians" and "Tune-Your-Style".
• [2025-03] 🔥 We release "NeuralGS", a novel framework that effectively adopts the neural field representation to encode the attributes of 3D Gaussians with multiple tiny MLPs.
• [2025-01] 🔥 We release "AE-NeRF", a novel framework for 3D reconstruction using event streams under noisy poses and in unbounded larger scenes.
• [2024-12] Two papers are accepted by AAAI 2025.
• [2024-05] 🔥 We release "EvaGaussians", the first approach that reconstructs sharp static 3D scenes with sparse RGB frames and event streams.

🗂️ Selected Projects

♠ Multimodal Generation 🤖 LingBot-Video: Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence 🎨 Style-GRPO: Semantic-Aware Preference Optimization for Image Style Transfer Guided by Reward Modeling 👹 Enhancing Spatial Understanding in Image Generation via Reward Modeling 🍟 WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation

🤖 LingBot-Video: Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence

Shuailei Ma, Jiaqi Liao, Xinyang Wang, Jingjing Wang, Chaoran Feng, Zijing Hu, Chong Bao, Zichen Xi, Yuqi Gan, Weisen Wang, Yanhong Zeng, Qin Zhao, Zifan Shi, Wei Wu, Hao Ouyang, Qiuyu Wang, Shangzhan Zhang, Jiahao Shao, Yipengjing Sun, Liangxiao Hu, Lunke Pan, Nan Xue, Kecheng Zheng, Yinghao Xu, Xing Zhu, Yujun Shen, Ka Leong Cheng

A DiT-based MoE video pretraining paradigm tailored for embodied intelligence. Contributed post-training reward design and the RL framework for physical rationality and task completion alignment.

Technical Report Paper Project Page Code ★629

🎨 Style-GRPO: Semantic-Aware Preference Optimization for Image Style Transfer Guided by Reward Modeling

Chaoran Feng^*,#, Jianbin Zhao*, Miao Yu* , Yingtao Li , Zhenyu Tang , Wangbo Yu , Yian Zhao , Xiaomin Li , Li Yuan† , Yonghong Tian†

A novel framework for image style transfer using semantic-aware preference optimization guided by reward modeling.

CVPR 2026 Paper

👹 Enhancing Spatial Understanding in Image Generation via Reward Modeling

Zhenyu Tang*, Chaoran Feng^*, Yufan Deng, Jie Wu, Xiaojie Li, Rui Wang, Yunpeng Chen, Daquan Zhou

A novel framework for enhancing spatial understanding in image generation via reward modeling.

CVPR 2026 Paper Project Page Code ★85

🍟 WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation

Yuwei Niu, Munan Ning, Mengren Zheng, Weiyang Jin, Bin Lin, Peng Jin, Jiaqi Liao, Chaoran Feng, Fanqing Meng, Kunpeng Ning, Bin Zhu, Li Yuan

The first benchmark for evaluating world-knowledge-informed semantic understanding in text-to-image generation, with 1,000 prompts across cultural common sense, spatio-temporal reasoning, and natural science.

ICML 2026 Paper Code ★210

♠ 3D Vision 👾 EA3D: Event-Augmented 3D Diffusion for Generalizable Novel View Synthesis 🌫️ DeblurNVS: Geometric Latent Diffusion for Novel View Synthesis from Sparse Motion-Blurred Images 🍿 Breaking the Vicious Cycle: Coherent 3D Gaussian Splatting from Sparse and Motion-Blurred Views 🎨 Tune-Your-Style: Intensity-tunable 3D Style Transfer with Gaussian Splatting 🔥 Cycle3D: High-quality and Consistent Image-to-3D Generation via Generation-Reconstruction Cycle 🌀 NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations 👤 NOFA++: Tuning-free NeRF-based One-shot Facial Avatar Reconstruction

👾 EA3D: Event-Augmented 3D Diffusion for Generalizable Novel View Synthesis

Wangbo Yu*, Chaoran Feng^*, Jianing Li, Aofan Zhang, Zhenyu Tang, Li Yuan† , Yonghong Tian†

An Event-Augmented 3D Diffusion framework for generalizable novel view synthesis from event streams and sparse RGB inputs.

ICLR 2026

🌫️ DeblurNVS: Geometric Latent Diffusion for Novel View Synthesis from Sparse Motion-Blurred Images

Changyue Shi, Wangbo Yu, Chaoran Feng, Li Yuan†

A geometric latent diffusion framework for high-fidelity novel view synthesis directly from sparse motion-blurred images without per-scene optimization.

arXiv 2026 Paper Code ★15

🍿 Breaking the Vicious Cycle: Coherent 3D Gaussian Splatting from Sparse and Motion-Blurred Views

Chaoran Feng, Zhankuo Xu, Yingtao Li, Jianbin Zhao, Jiashu Yang, Wangbo Yu, Li Yuan†, Yonghong Tian†

A novel framework for coherent 3D Gaussian Splatting from sparse and motion-blurred views using physics-aware deblurring priors coupled with diffusion-driven geometry completion.

IEEE T-CSVT Paper Code ★98

🎨 Tune-Your-Style: Intensity-tunable 3D Style Transfer with Gaussian Splatting

Yian Zhao, Rushi Ye, Ruochong Zheng, Zesen Cheng, Chaoran Feng, Jiashu Yang, Pengchong Qiao, Chang Liu, Jie Chen†

A novel style transfer framework with 3D Gaussian Splatting.

ICCV 2025 Paper

🔥 Cycle3D: High-quality and Consistent Image-to-3D Generation via Generation-Reconstruction Cycle

Zhenyu Tang*, Junwu Zhang*, Xinhua Cheng, Wangbo Yu, Chaoran Feng, Yatian Pang, Bin Lin, Li Yuan†

The project is about 3D generation using a generation-reconstruction cycle for a unified diffusion process.

AAAI 2025 Paper Code ★219

🌀 NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations

Zhenyu Tang*, Chaoran Feng*, Xinhua Cheng, Wangbo Yu, Junwu Zhang, Yuan Liu†, Xiaoxiao Long, Wenping Wang, Li Yuan†

A novel framework using neural fields to encode 3D Gaussians with compact MLPs for large-scale scenes.

AAAI 2026 Paper Code ★183

👤 NOFA++: Tuning-free NeRF-based One-shot Facial Avatar Reconstruction

Wangbo Yu, Chaoran Feng, Li Yuan†, and Yonghong Tian†

One-shot 3D facial avatar reconstruction with high fidelity and dynamic reenactment from a single image.

IEEE T-CSVT Paper

♠ Neuromorphic Vision 🍡 GS2E: Gaussian Splatting is an Effective Data Generator for Event Stream Generation 🎈 E-4DGS: High-Fidelity Dynamic Reconstruction from the Multi-view Event Cameras ✨ EvaGaussians: Event Assisted Gaussian Splatting from Blurry Images ⚡ AE-NeRF: Augmenting Event-Based Neural Radiance Fields for Non-ideal Conditions and Larger Scenes

🍡 GS2E: Gaussian Splatting is an Effective Data Generator for Event Stream Generation

Yuchen Li*, Chaoran Feng^*,#, Zhenyu Tang, Kaiyuan Deng, Wangbo Yu, Yonghong Tian†, Li Yuan†

The large-scale event dataset and a novel pipeline to simulate the event stream with 3DGS.

NeurIPS 2025 D&B Track Paper Code ★20 Datasets

🎈 E-4DGS: High-Fidelity Dynamic Reconstruction from the Multi-view Event Cameras

Chaoran Feng, Zhenyu Tang, Wangbo Yu, Yatian Pang, Yian Zhao, Jianbin Zhao, Li Yuan†, Yonghong Tian†

A novel framework to reconstruct high-fidelity scenes with fast-motion event cameras.

ACM MM 2025 Paper

✨ EvaGaussians: Event Assisted Gaussian Splatting from Blurry Images

Wangbo Yu*, Chaoran Feng*, Jiye Tang, Jiashu Yang, Zhenyu Tang, Xu Jia, Yuchao Yang, Li Yuan†, Yonghong Tian†

Event-assisted 3D reconstruction from blurry images with noisy poses and dynamic scenes.

ICCV 2025 Paper Code ★60

⚡ AE-NeRF: Augmenting Event-Based Neural Radiance Fields for Non-ideal Conditions and Larger Scenes

Chaoran Feng, Wangbo Yu, Xinhua Cheng, Zhenyu Tang, Junwu Zhang, Li Yuan†, Yonghong Tian†

3D reconstruction with event streams under noisy poses and unbounded scenes.

AAAI 2025 Paper Code

💼 Experience

Research Intern, Horizon Robotics

Mar. 2025 - Sep. 2025

Hosted by Yanfeng Zhao

Research Intern, Ant Research

Mar. 2026 - Now

Hosted by Ka Leong Cheng

🧑‍💻 Professional Activities

• Conference Reviewer: ICLR[25-26], CVPR[25-26], ICCV[25], NeurIPS[24-25], AAAI[25-26], IJCAI[25], ACM MM[25]
• Journal Reviewer: T-PAMI, T-CSVT