Education

Ph.D. - Computer Application Technology, East China Normal University and Shanghai Innovation Institute
Shanghai, 2025-2028 (expected), supervised by Prof. Wei Zhang
M.S. - Computer Science and Technology, the School of Computer Science and Technology, East China Normal University
Shanghai, 2022-2025, advised by Prof. Liang He and Prof. Qin Chen from ICALK
B.S. - Artificial Intelligence, the School of Computer Science and Technology, Soochow University
Suzhou, 2018-2022, guided by Prof. Lei Zhao from ADA LAB
High School - School Affiliated to Nanjing Normal University
Nanjing, 2014-2017

Work Experience

October 2024 - September 2025: Shanghai Artificial Intelligence Laboratory (Shanghai AI Lab)
- Position: full-time research intern
  cooperating closely with Dr. Qipeng Guo and Dr. Linyang Li
- Duty: reasoning data synthesis for LLMs, agent framework construction
- Work: backbone training of InternLM3 and Intern-S1, data synthesis framework CargoFlow construction, agent reasoning framework InternBootcamp (v1.0 and v2.0) build up
May 2024 - October 2024: RedBook
- Position: full-time business intern
  cooperating with my friends Yinqi Zhang and Siyu Yan
- Duty: note title generation, note title review, note content generation, research for hallucination detection
- Work: business-focused research in advertising algorithms
March 2023 - April 2023: Shanghai Institute of Intelligent Education
- Position: part-time research intern
- Duty: early-stage research on LLMs, expansion of the Chinese vocabulary, LLM pretraining
- Work: the first education LLM in China: Educhat (v1.0)

All publications can be found on my Google Scholar page here.

LLM Reasoning

A Survey of Inductive Reasoning for Large Language Models [arxiv] [github]
Kedi Chen, Dezhao Ruan, Yuhao Dan, Yaoting Wang, Siyu Yan, Xuecheng Wu, Yinqi Zhang, Qin Chen, Jie Zhou, Liang He, Biqing Qi, Linyang Li, Qipeng Guo, Xiaoming Shi, Wei Zhang†. In arxiv2025 (Long)
Code-driven Number Sequence Calculation: Enhancing the Inductive Reasoning Abilities of Large Language Models [arxiv] [github]
Kedi Chen, Zhikai Lei, Xu Guo, Xuecheng Wu, Siyuan Zeng, Jianghao Yin, Yinqi Zhang, Qin Chen, Jie Zhou, Liang He, Qipeng Guo, Kai Chen, Wei Zhang†. In arxiv2025 (Long)
InternBootcamp Technical Report: Boosting LLM Reasoning with Verifiable Task Scaling [arxiv] [github]
Peiji Li*, Jiasheng Ye*, Yongkang Chen*, Yichuan Ma, Zijie Yu, Kedi Chen, Ganqu Cui, Haozhan Li, Jiacheng Chen, Chengqi Lyu, Wenwei Zhang, Linyang Li†, Qipeng Guo, Dahua Lin, Bowen Zhou, Kai Chen†. In arxiv2025
Complete Chess Games Enable LLM Become A Chess Master [arxiv][github]
Yinqi Zhang, Xintian Han, Haolong Li, Kedi Chen, Shaohui Lin†. In NAACL2025 (Short)

LLM Backbone Training

Intern-S1: A Scientific Multimodal Foundation Model [arxiv] [github]
Intern-S1 Team, Shanghai AI Laboratory. In arxiv2025 (contribution of chemical synthesis data)

Agntic LLM

LLM Hallucination

HalluScope: A Comprehensive Dataset for Evaluating Hallucination in Large Language Models Across Multiple Domains
Chen Zhao, BiaoJie Zeng, Kedi Chen, Xin Lin†. In ICIC2025 (Long)
Enhancing Uncertainty Modeling with Semantic Graph for Hallucination Detection [arxiv] [github]
Kedi Chen*, Qin Chen*†, Jie Zhou, Xinqi Tao, Bowen Ding, Jingwen Xie, Mingchen Xie, Peilong Li, Feng Zheng. In AAAI2025 (Long)
DiaHalu: A Dialogue-level Hallucination Evaluation Benchmark for Large Language Models [arxiv] [github]
Kedi Chen, Qin Chen†, Jie Zhou, Yishen He, Liang He. In EMNLP2024 (findings, Long)

LLM Applications

Beyond Quality: Unlocking Diversity in Ad Headline Generation with Large Language Models [arxiv] [github]
Chang Wang*, Siyu Yan*, Depeng Yuan, Yuqi Chen, Yanhua Huang†, Yuanhang Zheng, Shuhao Li, Yinqi Zhang, Kedi Chen, Mingrui Zhu, Ruiwen Xu. In arxiv2025 (Long)

Old School

A Regularization-based Transfer Learning Method for Information Extraction via Instructed Graph Decoder [arxiv] [github]
Kedi Chen, Jie Zhou, Qin Chen†, Shunyu Liu, Liang He. In COLING2024 (Long)
Causal Intervention-based Prompt Debiasing for Event Argument Extraction [arxiv]
Jiaju Lin, Kedi Chen, Qin Chen†, Jie Zhou, Liang He. In arxiv2023 (Long)
Recommendation Model Based on Heterogeneous Information Networks
Kedi Chen, Lei Zhao†, Xinyi Chen, Kenan Shi. In Computer System Applications (2022.8) (Long, in Chinese)

*: equal contribution †: corresponding author

Language: Chinese, English
Programming language: Python (mostly used), C++ (used in algorithmic competitions)
Deep Learning Framework: Pytorch, FastNLP, Firefly, LLaMA-Factory, InternTrainer, Opnecompass, Cargoflow, Verl, RL-factory

reviewer of conferences: AAAI, PRCV, ACL ARR, COLING, CVPR, ICME, ICML, KDD, ICLR
student membership of the China Computer Federation
student membership of the Committee on Natural Language Processing in Shanghai
president of the Military Enthusiasts Association of Soochow University from 2019 to 2020