Here is Guoxin Chen (陈国鑫).

I am now a first-year PhD student at Renmin University of China, advised by Prof. Wayne Xin Zhao and Ruihua Song.

My research focuses primarily on LLM Reasoning and Agent. Currently, I have a strong research interest focusing on LLM Agent (particularly Search Agent and Code Agent). I am committed to enabling LLMs to possess human-like reasoning and tool-using capabilities.

Beyond research, I love traveling 🗺️ (highly recommend Shanxi Province for ancient architecture, Yili Loop in Xinjiang Province for stunning nature, and the Gansu-Qinghai Loop for epic landscapes!) and watching Crayon Shin-chan 🍜 (who can resist an episode during mealtime? ~).

If you are interested in any aspect of me, I would love to chat and collaborate, please email me at - gx.chen.chn [at] gmail [dot] com

🔥 News

• 2026.01: 🎉 Three Paper have been accepted at ICLR 2026, see you 🇧🇷～
  • IterResearch: A Markovian deep research framework enabling 2048+ interaction scaling with just 40K context length.
  • ReForm: Reflective autoformalization with iterative semantic verification and self-correction
• 2025.09: 📝 Tongyi DeepResearch and WebResearcher - Our technical report is now available! We present a complete pipeline for deep-research agents: from data synthesis (WebFrontier) to iterative paradigm (IterResearch) to test-time scaling. Individual papers on each component coming soon.
• 2025.05: 🎉 Table-Critic is accepted at ACL 2025 (Main). It’s my first paper as the corresponding author and guiding others to complete the work 😆😆😆. Thanks to all my collaborators!
• 2025.05: 🎉 C-3PO is accepted at ICML 2025. Thanks to my collaborators.
• 2025.01: 🎉 ToolEVO is accepted at ICLR 2025. Thanks to Zhong Zhang and Xin Cong for their guidance.
• 2024.09: 🎉 AlphaMath and DiReCT are accepted at NeurIPS 2024. Thanks to my collaborators.
• 2024.09: 🎉 SVPO is accepted at EMNLP-Findings 2024 (Well, findings again 😢).

📝 Publications

$^†$ is Corresponding author.

🌲 Tongyi DeepResearch Project

• Technical Report Tongyi DeepResearch Technical Report. Core Contributor.
• ICLR 2026 IterResearch: Rethinking Long-Horizon Agents with Interaction Scaling Guoxin Chen, Zile Qiao, Xuanzhong Chen, et al.
• Technical Report WebResearcher: Unleashing unbounded reasoning capability in Long-Horizon Agents. Core Contributor.
• ICLR 2026 AgentFrontier: Expanding the Capability Frontier of LLM Agents with ZPD-Guided Data Synthesis Xuanzhong Chen, Zile Qiao, Guoxin Chen, et al.

📚 Large Language Models

• ICLR 2026 ReForm: Reflective Autoformalization with Prospective Bounded Sequence Optimization Guoxin Chen, Jing Wu, et al.
• ACL 2025 Table-Critic: A Multi-Agent Framework for Collaborative Criticism and Refinement in Table Reasoning Peiying Yu, Guoxin Chen$^†$, Jingjing Wang.
• ICML 2025 C-3PO: Compact Plug-and-Play Proxy Optimization to Achieve Human-like Retrieval-Augmented Generation Guoxin Chen, Minpeng Liao, Peiying Yu, Dingmin Wang, Zile Qiao, Chao Yang, Wayne Xin Zhao, Kai Fan.
• ICLR 2025 Learning Evolving Tools for Large Language Models Guoxin Chen, Zhong Zhang, Xin Cong, et al.
• NeurIPS 2024 AlphaMath Almost Zero: process Supervision without process Guoxin Chen, Minpeng Liao, Chengxi Li, Kai Fan.
• NeurIPS 2024 DiReCT: Diagnostic Reasoning for Clinical Notes via Large Language Models Bowen Wang, Jiuyang Chang, Yiming Qian, Guoxin Chen, Junhao Chen, et al.
• EMNLP 2024 Step-level Value Preference Optimization for Mathematical Reasoning Guoxin Chen, Minpeng Liao, Chengxi Li, Kai Fan.
• ACL 2024 SEER: Facilitating Structured Reasoning and Explanation via Reinforcement Learning Guoxin Chen, Kexin Tang, Chao Yang, Fuying Ye, Yu Qiao, and Yiming Qian.
• EMNLP 2023 MPrompt: Exploring Multi-level Prompt Tuning for Machine Reading Comprehension, Guoxin Chen, Yiming Qian, Bowen Wang, and Liangzhi Li.

🌐 Graph Mining

• CIKM 2024 FCS-HGNN: Flexible Multi-type Community Search in Heterogeneous Information Networks Guoxin Chen, Fangda Guo, Yongqing Wang, et al.

• CIKM 2023 Causality and independence enhancement for biased node classification Guoxin Chen, Yongqing Wang, Fangda Guo, et al.

• CCIR 2022 InDNI: An Infection Time Independent Method for Diffusion Network Inference Guoxin Chen, Yongqing Wang, Jiangli Shao, et al.

📖 Educations

• 2025.09 - 2029.06 (Expected), Ph.D. student, Renmin University of China.
• 2022.09 - 2025.06, Master, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China.
• 2018.09 - 2022.06, Undergraduate, College of Science, Zhejiang University of Technology, Hangzhou, China.

💻 Internships

• 2025.12 - present: Research Intern at BandAI, ByteDance, Beijing, China

• 2024.03 - 2025.11: Research Intern at Tongyi Lab, Alibaba Group, Beijing, China

• 2023.09 - 2024.02: Research Intern at Shanghai AI Lab, Shanghai, China

–>