自我介绍

我目前在字节跳动Seed团队担任高级研究员，参与TopSeed人才计划。 我的研究方向专注于大语言模型（LLMs）的复杂推理能力与通用Agent基座模型。作为通用Agent优化的算法负责人之一，我参与了字节跳动Seed系列模型（如Seed 2.1-Seed 1.5）及豆包产品的研发，包括搭建豆包专业版办公任务模式（Office Task Mode），豆包专家模式和豆包超能模式。 我们正在招聘研究实习生并寻求学术合作，欢迎随时联系我：wanjun@bytedance.com

在此之前，我曾在华为诺亚方舟实验室担任高级研究员，参与华为天才少年计划。 我曾于2018至2023年期间，参与中山大学与微软亚洲研究院（MSRA）联合培养博士项目，并在中山大学计算机科学与工程学院获得博士学位。

作为联合培养博士生，我的导师包括周明博士、印鉴教授和王甲海教授。我曾在MSRA自然语言计算组实习，导师为段楠博士。

我于2021年获得微软学者奖学金（每年亚太地区11名杰出博士生），并于2023年入选华为天才少年计划及ACM广州优秀博士论文奖。

我曾在顶级AI会议和期刊上发表了50多篇论文，包括NeurIPS、ICLR、ACL、EMNLP、TASLP、NAACL、AAAI、IJCAI、ISSTA等。

🔥 News

• 2026.06 Release Seed 2.1 series, a next-generation agent foundation model for real-world productivity. Grateful to contribute to the general agent optimization as one of the algorithm core contributors.
• 2026.06 Released the Office Task Mode of 豆包专业版 (Doubao Pro), where I was fortunate to contribute as one of the algorithm leads.
• 2026.04 Released Agent-World (paper), exploring scalable real-world environment synthesis for evolving general agent intelligence. Check out our demo!
• 2026.02 Release Seed 2.0 as a core contributor of general agent (i.e., Multi-modal Agent, Tool-Learning Agent, Search Agent, etc.), the leading Agent foundation models!
• 2026.02 As the core contributor, released 豆包专家模式 (Doubao Expert Mode) — long-CoT reasoning that delivers expert-level answers to complex professional problems — and 豆包超能模式.
• 2025.12 Release Seed 1.8 as a core contributor.
• 2025.09 Release UI-TARS-2, a multi-modal unified agent models capable of coding, tool using and GUI operation, achieving leading performance.
• 2025.04: 🎉 Joined ByteDance Seed Edge team as Senior Research Scientist focusing on Large Language Models and Agent foundation models!
• 2025.04 Released ReTool: A reinforcement learning-based multi-turn tool-use agent training framework!
• 2025.04: 🎉 Seed-VL-v1.5 technical report released, advancing multi-modal models with great understanding and reasoning capabilities!
• 2025.04: 🎉 Seed-Thinking-v1.5 technical report released, advancing superb reasoning models with reinforcement learning
• 2025.01: 🎉 Released UI-TARS: Industry’s open-source GUI+Game Agent foundation model with 6.2K+ GitHub stars!
• 2024.06: 🎉 Joined ByteDance TopSeed program as a Senior Research Scientist!

📖 教育背景

• 2018.09 - 2023.06, 计算机科学与技术专业博士学位, 中山大学 (SYSU), 与微软亚洲研究院 (MSRA) 联合培养博士生项目
• 2014.09 - 2018.06, 软件工程专业学士学位, 中山大学数据科学与计算机学院 (SYSU)

💼 工作经历

• 2024.06 - 至今, 字节跳动Seed团队 - 大模型高级研究员
  • 职责：大语言模型和Agent方向高级研究员，参与TopSeed人才计划，通用Agent优化算法负责人之一
  • 项目经历：
    • Seed 1.8 & Seed 2.0 & Seed 2.1 (领先的新一代Agent基座模型，核心贡献者)
    • 豆包专业版办公任务模式 (Office Task Mode of Doubao Pro，算法负责人之一)
    • 豆包专家模式（基于长思维链推理，为专业复杂问题提供专家级深度解答，提升理科、代码、专业知识与创意写作等场景的作答质量）& 豆包超能模式
    • Agent-World (面向通用Agent智能进化的可扩展真实环境合成)
    • Seed-Thinking长思维链推理模型
    • Seed-Agent基座模型：
      • UI-TARS & UI-TARS-2 （业界领先的开源GUI+Game Agent基座模型）的训练
      • ReTool (Agent多轮工具调用强化学习训练框架)
      • MCP工具增强的DeepResearch模型及通用Agent基座模型训练
• 2023.06 - 2024.06, 华为诺亚方舟实验室 - 语音语义实验室 - 研究员（天才少年）
  • 项目经历：大语言模型方向研究员，专门负责盘古基础语言模型指令微调、数据飞轮、Agent超级对齐和复杂推理等研究和落地
• 2018.06 - 2023.06, 微软亚洲研究院 - 联合培养项目长期实习
  • 导师：段楠博士和周明博士

💬 学术指导

• 博士生导师：周明博士（澜舟科技CEO，前微软亚洲研究院副院长），印鉴教授（中山大学），王甲海教授（中山大学）
• 微软亚洲研究院导师：段楠博士（自然语言计算组）

🔬 研究实习

• 2018.06 - 2023.06, 研究实习生, 微软亚洲研究院 (MSRA) 自然语言计算组, 北京
  • 联合培养博士项目期间的长期实习
  • 导师：段楠博士

🎖 Honors and Awards

• 2024 字节跳动TopSeed人才计划 (ByteDance TopSeed)
• 2023 ACM中国-广州分会优秀博士论文奖 (ACM Outstanding Doctoral Thesis Award on China-Guangzhou)
• 2023 华为天才少年人才计划 (Huawei TopMinds)
• 2021 Microsoft Research Fellowship Award (11 outstanding Ph.D. students in computer science in the Asia-Pacific region each year)
• 2021 Baidu Scholarship (Global Top 40)
• 2021 National Scholarship of Ph.D., 2020 (Top 0.2%)
• 2016 The First Prize Scholarship

🏆 Competition Award

• 2023 Champion of CVPR 2023 Ego4D Challenge for Episodic Memory Natural Language Queries
• 2022 3rd of ECCV 2022 Ego4D Challenge for Episodic Memory Natural Language Queries
• 2018 Merit Awards of Global Artificial Intelligence Application Competition
• 2018 Rank 3rd, 7th in the quarter-finals of the 2018 FASHIONAI GLOBAL CHALLENGE

📝 Publications

Works in Seed

• Seed Technical Report Seed2.1: A Next-Generation Agent Foundation Model for Real-World Productivity
• arXiv 2026 Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence, Guanting Dong, Jianqiao Lu, Junjie Huang, Wanjun Zhong, Liang Liu, Shijue Huang, Zhenghao Li, Yang Zhao, Xiao Song, et al.
• Seed Technical Report Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity
• Seed Technical Report Seed1.8 Model Card: Towards Generalized Real-World Agency
• Seed Technical Report UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning
• Seed Technical Report Seed1.5-VL Technical Report
• Seed Technical Report Seed-Thinking-v1.5: Advancing Superb Reasoning Models with Reinforcement Learning
• Seed Technical Report UI-TARS: Pioneering Automated GUI Interaction with Native Agents
• Sub. to NeurIPS 2025 Retool: Reinforcement learning for strategic tool use in llms, Jiazhan Feng, Shijue Huang, Xingwei Qu, Ge Zhang, Yujia Qin, Baoquan Zhong, Chengquan Jiang, Jinxin Chi, Wanjun Zhong
• Arxiv Autokaggle: A multi-agent framework for autonomous data science competitions, Ziming Li, Qianbo Zang, David Ma, Jiawei Guo, Tuney Zheng, Minghao Liu, Xinyao Niu, Yue Wang, Jian Yang, Jiaheng Liu, Wanjun Zhong, Wangchunshu Zhou, Wenhao Huang, Ge Zhang
• EMNLP 2025 Otc: Optimal tool calls via reinforcement learning, Hongru Wang, Cheng Qian, Wanjun Zhong, Xiusi Chen, Jiahao Qiu, Shijue Huang, Bowen Jin, Mengdi Wang, Kam-Fai Wong, Heng Ji

Large Language Model Reasoning

• Seed Technical Report Seed-Thinking-v1.5: Advancing Superb Reasoning Models with Reinforcement Learning
• ICLR 2025 G-llava: Solving geometric problem with multi-modal large language model, Jiahui Gao, Renjie Pi, Jipeng Zhang, Jiacheng Ye, Wanjun Zhong, Yufei Wang, Lanqing Hong, Jianhua Han, Hang Xu, Zhenguo Li, Lingpeng Kong
• ACL 2025 Self-reasoning language models: Unfold hidden reasoning chains with few reasoning catalyst, Hongru Wang, Deng Cai, Wanjun Zhong, Shijue Huang, Jeff Z Pan, Zeming Liu, Kam-Fai Wong
• Information Processing & Management Adaptive-solver framework for dynamic strategy selection in large language model reasoning, Jianpeng Zhou, Wanjun Zhong, Yanlin Wang, Jiahai Wang
• AAAI 2025 Exploring iterative enhancement for improving learnersourced multiple-choice question explanations with large language models, Qiming Bao, Juho Leinonen, Alex Yuxuan Peng, Wanjun Zhong, Gaël Gendron, Timothy Pistotti, Alice Huang, Paul Denny, Michael Witbrock, Jiamou Liu

General Agent Model & System

Multi-modal Agent (GUI etc.)

• Seed Technical Report UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning
• arXiv 2025 Game-TARS: Pretrained Foundation Models for Scalable Generalist Multimodal Game Agents, Zihao Wang, Xu Li, Yining Ye, Junjie Fang, Haoming Wang, Liang Liu, Shihao Liang, Jianqiao Lu, Zhenghao Wu, Jiazhan Feng, …, Wanjun Zhong, et al.
• Seed Technical Report UI-TARS: Pioneering Automated GUI Interaction with Native Agents, Yujia Qin, Yining Ye, Junjie Fang, Haoming Wang, Shihao Liang, Shizuo Tian, Junda Zhang, Jiahao Li, Yunxin Li, Shijue Huang, Wanjun Zhong, Kuanye Li, Jiale Yang, Yu Miao, Woyu Lin, Longxiang Liu, Xu Jiang, Qianli Ma, Jingyu Li, Xiaojun Xiao, Kai Cai, Chuang Li, Yaowei Zheng, Chaolin Jin, Chen Li, Xiao Zhou, Minchao Wang, Haoli Chen, Zhaojian Li, Haihua Yang, Haifeng Liu, Feng Lin, Tao Peng, Xin Liu, Guang Shi

Tool-Learning Agent

• Sub. to NeurIPS 2025 Retool: Reinforcement learning for strategic tool use in llms, Jiazhan Feng, Shijue Huang, Xingwei Qu, Ge Zhang, Yujia Qin, Baoquan Zhong, Chengquan Jiang, Jinxin Chi, Wanjun Zhong
• Arxiv Autokaggle: A multi-agent framework for autonomous data science competitions, Ziming Li, Qianbo Zang, David Ma, Jiawei Guo, Tuney Zheng, Minghao Liu, Xinyao Niu, Yue Wang, Jian Yang, Jiaheng Liu, Wanjun Zhong, Wangchunshu Zhou, Wenhao Huang, Ge Zhang
• EMNLP 2025 Otc: Optimal tool calls via reinforcement learning, Hongru Wang, Cheng Qian, Wanjun Zhong, Xiusi Chen, Jiahao Qiu, Shijue Huang, Bowen Jin, Mengdi Wang, Kam-Fai Wong, Heng Ji
• ACL 2024 Planning, Creation, Usage: Benchmarking LLMs for Comprehensive Tool Utilization in Real-World Complex Scenarios, Shijue Huang, Wanjun Zhong, Jianqiao Lu, Qi Zhu, Jiahui Gao, Weiwen Liu, Yutai Hou, Xingshan Zeng, Yasheng Wang, Lifeng Shang, Xin Jiang, Ruifeng Xu, Qun Liu

Code Agent

• Arxiv Agents in software engineering: Survey, landscape, and vision, Yanlin Wang, Wanjun Zhong, Yanxian Huang, Ensheng Shi, Min Yang, Jiachi Chen, Hui Li, Yuchi Ma, Qianxiang Wang, Zibin Zheng
• ICSME 2023 You Augment Me: Exploring ChatGPT-based Data Augmentation for Semantic Code Search, Yanlin Wang, Lianghong Guo, Ensheng Shi, Wenqing Chen, Jiachi Chen, Wanjun Zhong, Menghan Wang, Hui Li, Hongyu Zhang, Ziyu Lyu, Zibin Zheng
• ISSTA 24 - Outstanding Paper Award When to stop? towards efficient code generation in llms with excess token prevention, Lianghong Guo, Yanlin Wang, Ensheng Shi, Wanjun Zhong, Hongyu Zhang, Jiachi Chen, Ruikai Zhang, Yuchi Ma, Zibin Zheng

Agent Memory

• AAAI 2024 MemoryBank: Enhancing Large Language Models with Long-Term Memory, Wanjun Zhong, Lianghong Guo, Qiqi Gao, He Ye, Yanlin Wang

Agent-driven Training

• arXiv 2026 Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence, Guanting Dong, Jianqiao Lu, Junjie Huang, Wanjun Zhong, Liang Liu, Shijue Huang, Zhenghao Li, Yang Zhao, Xiao Song, et al.
• arXiv 2024 YODA: Teacher-Student Progressive Learning for Language Models, Jianqiao Lu, Wanjun Zhong, Yufei Wang, Zhijiang Guo, Qi Zhu, Wenyong Huang, Yanlin Wang, Fei Mi, Baojun Wang, Yasheng Wang, Lifeng Shang, Xin Jiang, Qun Liu (* equal contribution)

Benchmark and Evaluation

• NAACL 2024 AGIEval: A Human-Centric Benchmark for Evaluating Foundation Models, Wanjun Zhong, Ruixiang Cui, Yiduo Guo, Yaobo Liang, Shuai Lu, Yanlin Wang, Amin Saied, Weizhu Chen, Nan Duan
• EMNLP 2024 CLongEval: A Chinese Benchmark for Evaluating Long-Context Large Language Models, Zexuan Qiu, Jingjing Li, Shijue Huang, Wanjun Zhong, Irwin King
• ACL 2024 Workshop PerLTQA: A Personal Long-Term Memory Dataset for Memory Classification, Retrieval, and Synthesis in Question Answering, Yiming Du, Hongru Wang, Zhengyi Zhao, Bin Liang, Baojun Wang, Wanjun Zhong, Zezhong Wang, Kam-Fai Wong
• ACL 2024 Followbench: A multi-level fine-grained constraints following benchmark for large language models, Yuxin Jiang, Yufei Wang, Xingshan Zeng, Wanjun Zhong, Liangyou Li, Fei Mi, Lifeng Shang, Xin Jiang, Qun Liu, Wei Wang

Self-Learning of LLMs

• AAAI 2025 Empowering Self-Learning of LLMs: Inner Knowledge Explicitation as a Catalyst, Shijue Huang, Wanjun Zhong, Deng Cai, Fanqi Wan, Chengyi Wang, Mingxuan Wang, Mu Qiao, Ruifeng Xu
• arXiv 2023 SELF: Language-driven self-evolution for large language model, Jianqiao Lu, Wanjun Zhong, Wenyong Huang, Yufei Wang, Fei Mi, Baojun Wang, Weichao Wang, Lifeng Shang, Qun Liu

General LLM Training

• arXiv 2023 Data management for large language models: A survey, Zige Wang, Wanjun Zhong, Yufei Wang, Qi Zhu, Fei Mi, Baojun Wang, Lifeng Shang, Xin Jiang, Qun Liu
• arXiv 2023 Aligning large language models with human: A survey, Yufei Wang, Wanjun Zhong, Liangyou Li, Fei Mi, Xingshan Zeng, Wenyong Huang, Lifeng Shang, Xin Jiang, Qun Liu
• ACL 2024 Learning to Edit: Aligning LLMs with Knowledge Editing, Yuxin Jiang, Yufei Wang, Chuhan Wu, Wanjun Zhong, Xingshan Zeng, Jiahui Gao, Liangyou Li, Xin Jiang, Lifeng Shang, Ruiming Tang, Qun Liu, Wei Wang

• ICSME 2023 You Augment Me: Exploring ChatGPT-based Data Augmentation for Semantic Code Search, Yanlin Wang, Lianghong Guo, Ensheng Shi, Wenqing Chen, Jiachi Chen, Wanjun Zhong, Menghan Wang, Hui Li, Hongyu Zhang, Ziyu Lyu, Zibin Zheng

• ACL 2023 CONE: An Efficient COarse-to-fiNE Alignment Framework for Long Video Temporal Grounding, Zhijian Hou, Wanjun Zhong, Leiji, Kun Yan, Difei Gao, Wing-Kwong Chan, Chong-Wah Ngo, Zheng Shou, Nan Duan (* equal contribution)

Previous Work Before 2023

Multi-Modal

• ICME 2023 Semantic Composition and Alignment with Cross-Modality-Aware Syntactic Hypergraph Convolutional Network for Video Question Answering, Zenan Xu, Wanjun Zhong, Qinliang Su, Zijing Ou, Fuwei Zhang (* equal contribution)
• ECCV Challenge 2022 An Efficient COarse-to-fiNE Alignment Framework @ Ego4D Natural Language Queries Challenge 2022, Zhijian Hou, Wanjun Zhong, Leiji, Kun Yan, Difei Gao, Wing-Kwong Chan, Chong-Wah Ngo, Zheng Shou, Nan Duan (* equal contribution) [code]

Knowledge-enhanced Language Model Reasoning

• ACL 2022 Disentangling Reasoning Capabilities from Language Models with Compositional Reasoning Transformers, Wanjun Zhong, Tingting Ma, Jiahai Wang, Jian Yin, Tiejun Zhao, Chin-Yew Lin, Nan Duan (* equal contribution)
• EMNLP 2022 Mixed-modality Representation Learning and Pre-training for Joint Table-and-Text Retrieval in OpenQA, JunJie Huang, Wanjun Zhong, Qian Liu, Ming Gong, Daxin Jiang, Nan Duan ( equal contribution) [code]
• NeurIPS 2022 LogiGAN: Learning Logical Reasoning via Adversarial Pre-training, Xinyu Pi, Wanjun Zhong, Yan Gao, Jian-guang Lou, Nan Duan (* equal contribution) [code]
• IJCAI 2022 (Oral) Reasoning over Hybrid Chain for Table-and-Text Open Domain Question Answering, Wanjun Zhong, Junjie Huang, Qian Liu, Ming Zhou, Jiahai Wang, Jian Yin, Nan Duan (* equal contribution) [code]
• NAACL 2022 ProQA: Structural Prompt-based Pre-training for Unified Question Answering, Wanjun Zhong, Yifan Gao, Ning Ding, Yujia Qin, Zhiyuan Liu, Ming Zhou, Jiahai Wang, Jian Yin, Nan Duan ( equal contribution) [code]
• TASLP 2022 From LSAT: The Progress and Challenges of Complex Reasoning, Siyuan Wang, Zhongkun Liu, Wanjun Zhong, Ming Zhou, Zhongyu Wei, Zhumin Chen, Nan Duan
• NAACL 2022 Analytical Reasoning of Text, Wanjun Zhong, Siyuan Wang, Duyu Tang, Zenan Xu, Daya Guo, Jiahai Wang, Jian Yin, Ming Zhou, Nan Duan [code]
• ACL 2022 Logic-Driven Context Extension and Data Augmentation for Logical Reasoning of Text, Siyuan Wang, Wanjun Zhong, Duyu Tang, Zhongyu Wei, Zhihao Fan, Daxin Jiang, Ming Zhou, Nan Duan [code]
  Here is the formatted list in the requested style:
• EMNLP 2021 WhiteningBERT: An Easy Unsupervised Sentence Embedding Approach, Junjie Huang, Duyu Tang, Wanjun Zhong, Shuai Lu, Linjun Shou, Ming Gong, Daxin Jiang, Nan Duan
• ACL 2021 UserAdapter: Few-Shot User Learning in Sentiment Analysis, Wanjun Zhong, Duyu Tang, Jiahai Wang, Jian Yin, Nan Duan
• ACL 2021 Syntax-Enhanced Pre-trained Model, Zenan Xu, Daya Guo, Duyu Tang, Qinliang Su, Linjun Shou, Ming Gong, Wanjun Zhong, Xiaojun Quan, Daxin Jiang, Nan Duan
• ACL 2021 Compare to The Knowledge: Graph Neural Fake News Detection with External Knowledge, Linmei Hu, Tianchi Yang, Luhao Zhang, Wanjun Zhong, Duyu Tang, Chuan Shi, Nan Duan, Ming Zhou
• EMNLP 2020 (Oral) Neural Deepfake Detection with Factual Structure of Text, Wanjun Zhong, Duyu Tang, Zenan Xu, Ruize Wang, Nan Duan, Ming Zhou, Jiahai Wang, Jian Yin [video]
• EMNLP 2020 Leveraging declarative knowledge in text and first-order logic for fine-grained propaganda detection, Ruize Wang, Duyu Tang, Nan Duan, Wanjun Zhong, Zhongyu Wei, Xuanjing Huang, Daxin Jiang, Ming Zhou
• ACL 2020 LogicalFactChecker: Leveraging Logical Operations for Fact Checking with Graph Module Network, Wanjun Zhong, Duyu Tang, Zhangyin Feng, Nan Duan, Ming Zhou, Ming Gong, Linjun Shou, Daxin Jiang, Jiahai Wang, Jian Yin [video]
• ACL 2020 Reasoning Over Semantic-Level Graph for Fact Checking, Wanjun Zhong, Jingjing Xu, Duyu Tang, Zenan Xu, Nan Duan, Ming Zhou, Jiahai Wang, Jian Yin [video]
• NLPCC 2019 Improving Question Answering by Commonsense-Based Pre-Training, Wanjun Zhong, Duyu Tang, Nan Duan, Ming Zhou, Jiahai Wang, Jian Yin
• AI Open 2023 Improving Task Generalization via Unified Schema Prompt, Wanjun Zhong, Yifan Gao, Ning Ding, Zhiyuan Liu, Ming Zhou, Jiahai Wang, Jian Yin, Nan Duan
• arXiv 2020 A Heterogeneous Graph with Factual, Temporal and Logical Knowledge for Question Answering Over Dynamic Contexts, Wanjun Zhong, Duyu Tang, Nan Duan, Ming Zhou, Jiahai Wang, Jian Yin