宋子帆

I am a Ph.D candidate in Computer Science and Technology at Tongji University, advised by Prof. Cairong Zhao and Guosheng Hu. Prior to this, I obtained my bachelor degree in Computer Science and Technology from Tongji University. My research interests lie in Multimodal Learning, Data-centric AI, and LLM Agent.


Education



Tongji University
Doctoral Student
Computer Science and Technology
2021.09 - Present
Tongji University
Bachelor Degree
Computer Science and Technology
2017.09 - 2021.06


Internship



Microsoft Research Asia
AI/ML Group
LLM Agent & World Model
2025.03 - 2025.12
Shanghai AI Laboratory
OpenMMLab @InternLM
LLM Fine-tuning & Code LLM
2023.09 - 2024.06


News

  • [2026/01] 2 Papers on Agentic System | Multimodal Dataset Distillation is accepted by the Fourteenth International Conference on Learning Representations (ICLR), 2026.
  • [2025/09] 1 Paper on RAG for LLMs is accepted by the 40th Annual AAAI Conference on Artificial Intelligence (AAAI), 2026.
  • [2025/09] 1 Paper on Learning Paradigm for LLMs is accepted by Advances in Neural Information Processing Systems (NeurIPS), 2025.
  • [2025/09] I have been invited to serve as a Reviewer for International Conference on Learning Representations (ICLR), 2026.
  • [2025/08] I have been invited to serve as a Reviewer for IEEE Transactions on Image Processing (TIP).
  • [2025/07] 1 Paper on Training Paradigm is accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025.


Selected Publications

Rationality-Driven Agentic System
Zifan Song, Kaitao Song, Guosheng Hu, Ding Qi, Junyao Gao, Xiaohua Wang, Dongsheng Li, Cairong Zhao
International Conference on Learning Representations (ICLR), 2026
[📃 Paper]


DPL++: Advancing the Network Performance via Image and Label Perturbations
Zifan Song, Xiao Gong, Guosheng Hu, Shuguang Dou, Qingsong Zhao, Cairong Zhao
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025
[📃 Paper] [👨‍💻 Code]


AlchemistCoder: Harmonizing and Eliciting Code Capability by Hindsight Tuning on Multi-source Data
Zifan Song, Yudong Wang, Wenwei Zhang, Kuikun Liu, Chengqi Lyu, Demin Song, Qipeng Guo, Hang Yan, Dahua Lin, Kai Chen, Cairong Zhao
Advances in Neural Information Processing Systems (NeurIPS), 2024
[📃 Paper] [👨‍💻 Code] [🌐 Project Page] [🤗 HuggingFace]


InternLM2 Technical Report
InternLM Group
arXiv, 2024
[📃 Paper] [👨‍💻 Code] [🌐 Website]


Diverse Person: Customize Your Own Dataset for Text-Based Person Search
Zifan Song, Guosheng Hu, Cairong Zhao
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2024
[📃 Paper] [👨‍💻 Code]


Deep Perturbation Learning: Enhancing the Network Performance via Image Perturbations
Zifan Song, Xiao Gong, Guosheng Hu, Cairong Zhao
International Conference on Machine Learning (ICML), 2023
[📃 Paper] [👨‍💻 Code]


Learning Scene-Pedestrian Graph for End-to-End Person Search
Zifan Song, Cairong Zhao, Guosheng Hu, Duoqian Miao
IEEE Transactions on Industrial Informatics (TII), 2023
[📃 Paper] [👨‍💻 Code]


Honors & Services

  • National Scholarship for Doctoral Students, 2025
  • Talk on Data-Centric Collaborative Optimization of Features and Labels @ CSIG Young Scientists Conference, 2025
  • Conference Reviewer: ICLR, NeurIPS, ICML, AAAI, CVPR, ICCV, ACL, CAI
  • Journal Reviewer: TIP, IJCV, TMM


Friends

Shuguang Dou, Junyao Gao, Zefan Qu, Yuanpeng Tu, Yubin Wang, Siyuan Li, Yudong Wang, Wenwei Zhang