Xu Guo | Fudan University

I am Xu Guo (郭旭), a third-year PhD student at the School of Computer Science, Fudan University, working on large language models and data-centric artificial intelligence.

I believe that training data, including synthetic data, is a fundamental driver of modern AI systems. My research focuses on data-centric methods for language model training, spanning pre-training, post-training data synthesis, and agentic training. My recent first-author work explores robust pre-pre-training under noisy data, data and reward design for reinforcement learning with verifiable rewards, and instruction-following post-training.

This official academic homepage collects my publications and research updates. You can also find my work and citation record on Xu Guo’s Google Scholar profile, with additional publication and identity records on DBLP, ORCID, and GitHub. You can reach me at guox24@m.fudan.edu.cn.

news

Jul 03, 2026	Our new preprint, When Does Generating More Help? Disentangling Fixed-Source Synthesis from Source Expansion in Synthetic Data Scaling, is now available on arXiv.
Jul 02, 2026	I launched my personal academic homepage.

selected publications

When Does Generating More Help? Disentangling Fixed-Source Synthesis from Source Expansion in Synthetic Data Scaling

Xu Guo, Jian Tong, Zhihui Lu, and Qipeng Guo

arXiv preprint arXiv:2607.01727, 2026

arXiv Bib HTML PDF

@article{guo2026when,
  title = {When Does Generating More Help? Disentangling {Fixed-Source Synthesis} from {Source Expansion} in {Synthetic Data Scaling}},
  author = {Guo, Xu and Tong, Jian and Lu, Zhihui and Guo, Qipeng},
  journal = {arXiv preprint arXiv:2607.01727},
  year = {2026},
}

Synthetic Pre-Pre-Training Improves Language Model Robustness to Noisy Pre-Training Data

Xu Guo, Runyu Peng, Jian Tong, Yunhua Zhou, Haijun Lv, Zhihui Lu, and Qipeng Guo

arXiv preprint arXiv:2605.10129, 2026

arXiv Bib HTML

@article{guo2026synthetic,
  title = {Synthetic Pre-Pre-Training Improves Language Model Robustness to Noisy Pre-Training Data},
  author = {Guo, Xu and Peng, Runyu and Tong, Jian and Zhou, Yunhua and Lv, Haijun and Lu, Zhihui and Guo, Qipeng},
  journal = {arXiv preprint arXiv:2605.10129},
  year = {2026},
}

Rethinking Multiple-Choice Questions for RLVR: Unlocking Potential via Distractor Design

Xu Guo, Qiming Ge, Jian Tong, Kedi Chen, Jin Zhang, Xiaogui Yang, Xuan Gao, Haijun Lv, Zhihui Lu, Yicheng Zou, and 1 more author

In Findings of the Association for Computational Linguistics: ACL 2026, 2026

arXiv Bib HTML

@inproceedings{guo2026rethinking,
  title = {Rethinking Multiple-Choice Questions for {RLVR}: Unlocking Potential via Distractor Design},
  author = {Guo, Xu and Ge, Qiming and Tong, Jian and Chen, Kedi and Zhang, Jin and Yang, Xiaogui and Gao, Xuan and Lv, Haijun and Lu, Zhihui and Zou, Yicheng and Guo, Qipeng},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
}

IFDECORATOR: Wrapping Instruction Following Reinforcement Learning with Verifiable Rewards

Xu Guo, Tianyi Liang, Tong Jian, Xiaogui Yang, Ling-I Wu, Chenhui Li, Zhihui Lu, Qipeng Guo, and Kai Chen

arXiv preprint arXiv:2508.04632, 2025

arXiv Bib HTML

@article{guo2025ifdecorator,
  title = {{IFDECORATOR}: Wrapping Instruction Following Reinforcement Learning with Verifiable Rewards},
  author = {Guo, Xu and Liang, Tianyi and Jian, Tong and Yang, Xiaogui and Wu, Ling-I and Li, Chenhui and Lu, Zhihui and Guo, Qipeng and Chen, Kai},
  journal = {arXiv preprint arXiv:2508.04632},
  year = {2025},
}

Towards transferable adversarial attacks on vision transformers for image classification

Xu Guo, Peng Chen, Zhihui Lu, Hongfeng Chai, Xin Du, and Xudong Wu

Journal of Systems Architecture, 2024

DOI Bib HTML

@article{guo2024towards,
  title = {Towards transferable adversarial attacks on vision transformers for image classification},
  author = {Guo, Xu and Chen, Peng and Lu, Zhihui and Chai, Hongfeng and Du, Xin and Wu, Xudong},
  journal = {Journal of Systems Architecture},
  volume = {152},
  pages = {103155},
  year = {2024},
  doi = {10.1016/j.sysarc.2024.103155},
}