Chuhao Jin is a Ph.D. student at the Gaoling School of Artificial Intelligence, Renmin University of China (RUC GSAI), advised by Prof. Ruihua Song. His research centers on Interactive Digital Humans and Expressive Interaction, spanning persuasive dialogue, full-body motion generation, multimodal interaction, and embodied AI agents.

I build expressive digital humans that can talk, move, emote, and interact in real time. My recent work focuses on expressive 3D avatars, text-to-motion generation, multimodal agents, and AI characters that can communicate with people in emotionally and socially meaningful ways.

Beyond academic research, I have worked on agentic AI products and interactive digital human systems from prototype to open-source release, with experience in both large-model research teams and early-stage startup environments.

Research Interests: Interactive Digital Humans · Expressive Interaction · Multimodal Agents · Motion Generation · Persuasive Dialogue

💻 Experience

• 2025.03 - 2026.05, Co-founder and CTO, SentiPulse (思维光谱), China. Built emotionally expressive AI agents and interactive digital human products; led the development and open-source release of SentiAvatar.
• 2024.05 - 2025.02, Large Model Algorithm Intern, Kuaishou, KwaiYii Large Model Research Center (快意大模型研究中心), China.
• 2023.03 - 2023.09, Research Intern, Microsoft Research Asia (MSRA), China.
• 2021.09 - 2023.01, Research Intern, Meituan, China.
• 2020.12 - 2021.09, Research Intern, Beijing Academy of Artificial Intelligence (BAAI), China.

🛠️ Open Source & Products

• Lead author of SentiAvatar , an expressive and interactive 3D digital human framework with synchronized speech, motion, and facial expressions.
• Former Co-founder and CTO of SentiPulse (思维光谱), where I worked on emotionally expressive AI agents and interactive digital human products.
• Author of Text2Poster , a framework for automatically generating posters based on text.
• Author of BriVL-BUA-applications , a project of using the BriVL model for Chinese image-text retrieval.
• Author of LSTMForSentAnalysis , a tiny tutorial on LSTM-based text classification.
• Contributed code to BAAI-WuDao/BriVL .

🔥 News

• 2026.05:  🎉 SentiPulse (思维光谱) was selected by QbitAI (量子位) as one of the AIGC companies to watch in 2026.
• 2026.04:  🎉 We open-sourced SentiAvatar on April 8, an expressive and interactive 3D digital human framework.
• 2026.03:  🎉 One paper, PyraMotion, is accepted by ICME 2026!
• 2025.06:  🎉 We released PlanMoGPT, a progressive planning framework for text-to-motion synthesis.
• 2024.05:  🎉 One paper is accepted by ACL 2024!
• 2023.09:  🎉 One paper is accepted by EMNLP 2023!
• 2023.06:  🎉 I obtain my Master’s degree from RUC!

📝 Selected Publications

SentiAvatar: Towards Expressive and Interactive Digital Humans, 2026. [Project] [Code]

Chuhao Jin, Rui Zhang, Qingzhe Gao, Haoyu Shi, Dayu Wu, Yichen Jiang, Yihan Wu, Ruihua Song

An expressive 3D digital human framework with speech, facial expression, and body motion generation for real-time interaction.

PyraMotion: Coarse-to-Fine Full-Body Motion Generation with Body-Hand Coherence, ICME 2026.

*Chuhao Jin, *Rui Zhang, Haosen Li, Ruihua Song

A coarse-to-fine framework for full-body motion generation with improved body-hand coherence.

PlanMoGPT: Flow-Enhanced Progressive Planning for Text to Motion Synthesis, 2025.

Chuhao Jin, Haosen Li, Bingzi Zhang, Che Liu, Xiting Wang, Ruihua Song, Wenbing Huang, Ying Qin, Fuzheng Zhang, Di Zhang

A progressive planning framework for text-to-motion synthesis with flow-enhanced motion tokenization.

(Nominated by Area Chair for Best Paper) Persuading across Diverse Domains: a Dataset and Persuasion Large Language Model, ACL 2024.

*Chuhao Jin, *Kening Ren, Lingzhen Kong, Xiting Wang, Ruihua Song, Huan Chen

A cross-domain persuasion dataset and large language model for persuasive dialogue.

Joint Semantic and Strategy Matching for Persuasive Dialogue, EMNLP findings 2023.

Chuhao Jin, Yutao Zhu, Lingzhen Kong, Shijie Li, Xiao Zhang, Ruihua Song, Xu Chen, Huan Chen, Yuchong Sun, Yu Chen, Jun Xu

• Space Roaming and Imagination Production: Mediating Construction of Internet-celebrity Cities in Online Pictures, Journalism & Communication, 2023.30(05). Shengjun Jin, Jiangmei Li, Yuhao Li , Chuhao Jin (Corresponding Author)

• Text2Poster: Laying out Stylized Texts on Retrieved Images, ICASSP 2022. Chuhao Jin, Hongteng Xu, Ruihua Song and Zhiwu Lu

* indicates equal contribution.

🖨️ Other Preprints

• Pave the Way to Grasp Anything: Transferring Foundation Models for Universal Pick-Place Robots, *Jiange Yang, *Wenhui Tan, *Chuhao Jin, Bei Liu, Jianlong Fu, Ruihua Song, Limin Wang

• AlphaBlock: Embodied Finetuning for Vision-Language Reasoning in Robot Manipulation, *Chuhao Jin, *Wenhui Tan, *Jiange Yang, Bei Liu, Ruihua Song, Limin Wang, Jianlong Fu

• WenLan: Bridging Vision and Language by Large-Scale Multi-Modal Pre-Training, Wenlan Team of Wudao, BAAI

📖 Education

• 2023.09 - Now, Ph.D. Student, Gaoling School of Artificial Intelligence (GSAI), Renmin University of China, Beijing.
• 2020.09 - 2023.06, Master’s Student, Gaoling School of Artificial Intelligence (GSAI), Renmin University of China, Beijing.
• 2016.09 - 2020.06, Undergraduate Student, School of Computer Science and Technology, Harbin Institute of Technology, Weihai.