About me

Daocheng Fu is a Ph.D. candidate at Fudan University’s College of Computer Science and Artificial Intelligence, in a joint training program with Shanghai AI Laboratory. He is advised by Prof. Yu Qiao and Dr. Botian Shi, and his research focuses on data synthesis, language model reasoning and continual learning.

News

• [2026-01-13] TraineeBench establishes a dynamic evaluation framework that simulates realistic workplace scenarios to rigorously assess agents on context-aware scheduling for streaming tasks, active exploration under uncertainty, and continuous learning from experience.
• [2026-01-08] SGVR introduces a reinforcement learning framework that replaces sparse outcome-based supervision with dense, verifiable sub-goal rewards derived from formal proof skeletons. By guiding models through rigorous intermediate milestones, it significantly enhances geometric reasoning performance while demonstrating robust generalization to general mathematics and logical reasoning tasks.
• [2025-12-25] SymDrive proposes a unified diffusion-based framework that achieves both photorealistic novel-view synthesis and realistic traffic editing by leveraging a Symmetric Auto-regressive Online Restoration paradigm to recover fine-grained details from symmetric views. Furthermore, it utilizes this restoration capability to enable a training-free harmonization mechanism for seamless, lighting-consistent 3D vehicle insertion into driving scenes.
• [2025-09-30] RE-Searcher investigates the relationship between search agents and the search environment, proposing a solution to enhance the agent’s stability.
• [2025-08-30] We release the GeoTrust-train & GeoTrust-test dataset at huggingface.
• [2025-08-29] TrustGeoGen is a data engine for multimodal geometric reasoning, built upon a formal engine, designed to synthesize challenging, full-modality, and trustworthy geometric reasoning data. The introduction of Connection Thinking and Thinking Template enhances the geometric reasoning capabilities of Large Multimodal Models (LMMs) without compromising their general abilities.

Selected Publications

You can also find my articles on my Google Scholar profile.

Agents and Environments

• The Agent’s First Day: Benchmarking Learning, Exploration, and Scheduling in the Workplace Scenarios

  Daocheng Fu*, Jianbiao Mei*, Rong Wu*, Xuemeng Yang*, Jia Xu, Ding Wang, Pinlong Cai, Yong Liu, Licheng Wen, Botian Shi

Search Agent

• RE-Searcher: Robust Agentic Search with Goal-oriented Planning and Self-reflection

  Daocheng Fu*, Jianbiao Mei*, Licheng Wen, Xuemeng Yang, Cheng Yang, Rong Wu, Tao Hu, Siqi Li, Yufan Shen, Xinyu Cai, Pinlong Cai, Botian Shi, Yong Liu, Yu Qiao

• \(O^2\)-Searcher: A Searching-based Agent Model for Open-Domain Open-Ended Question Answering

  Jianbiao Mei*, Tao Hu*, Daocheng Fu*, Licheng Wen, Xuemeng Yang, Rong Wu, Pinlong Cai, Xinyu Cai, Xing Gao, Yu Yang, Chengjun Xie, Botian Shi, Yong Liu, Yu Qiao

Multi-modal Reasoning

• TrustGeoGen: Formal-Verified Data Engine for Trustworthy Multi-modal Geometric Problem Solving

  Daocheng Fu*, Jianlong Chen*, Renqiu Xia*, Zijun Chen*, Qi Liu, Yuan Feng, Hongbin Zhou, Renrui Zhang, Shiyang Feng, Peng Gao, Hongyuan Zha, Junchi Yan, Botian Shi, Yu Qiao, Bo Zhang

• Milestones over Outcome: Unlocking Geometric Reasoning with Sub-Goal Verifiable Reward

  Jianlong Chen, Daocheng Fu, Shengze Xu, Jiawei Chen, Yuan Feng, Yue Yang, Junchi Yan, Hongyuan Zha, Renqiu Xia

Autonomouse Driving

• Drive Like a Human: Rethinking Autonomous Driving with Large Language Models

  Daocheng Fu*, Licheng Wen*, Xin Li*, Min Dou, Pinlong Cai, Botian Shi, Yu Qiao

• Dilu: A knowledge-driven approach to autonomous driving with large language models

  Licheng Wen*, Daocheng Fu*, Xin Li*, Xinyu Cai, Tao Ma, Pinlong Cai, Min Dou, Botian Shi, Liang He, Yu Qiao

• LimSim: A long-term interactive multi-scenario traffic simulator

  Licheng Wen*, Daocheng Fu*, Song Mao, Pinlong Cai, Min Dou, Yikang Li, Yu Qiao

• Limsim++: A closed-loop platform for deploying multimodal llms in autonomous driving

  Daocheng Fu*, Wenjie Lei*, Licheng Wen, Pinlong Cai, Song Mao, Min Dou, Botian Shi, Yu Qiao

• SymDrive: Realistic and Controllable Driving Simulator via Symmetric Auto-regressive Online Restoration

  Zhiyuan Liu*, Daocheng Fu*, Pinlong Cai, Lening Wang, Ying Liu, Yilong Ren, Botian Shi, Jianqiang Wang

Traffic & Transportation

• Incremental path planning: Reservation system in V2X environment

  Daocheng Fu, Pinlong Cai, Yilun Lin, Song Mao, Licheng Wen, Yikang Li

• Trafficgpt: Viewing, processing and interacting with traffic foundation models

  Siyao Zhang*, Daocheng Fu*, Wenzhe Liang, Zhao Zhang, Bin Yu, Pinlong Cai, Baozhen Yao