Jian Guan 关 健

Jian Guan currently serves as a researcher at Ant Group, working with Wei Wu. He obtained his bachelor and Ph.D. degrees from Tsinghua University, advised by Minlie Huang. Jian has also interned at various institutions, including the Allen Institute for Artificial Intelligence in Seattle with the supervision of Hao Peng and Jesse Dodge (Jan.~Jun., 2023) and the University of Virginia with the supervision of Hongning Wang (Jul.~Sep., 2018).

His research interests include:

• Open-ended NLG and Evaluation [must-read paper list]
• Efficient Foundation Models
• LLM-based Complex Reasoning
• Personalized Alignment with Human Prefernece [survey]

🚀 🚀 🚀 We are currently seeking self-motivated interns to join our team at the Ant Group. If you are interested in working on cutting-edge research in the above research fileds, we encourage you to contact us. Our institute offers a dynamic and collaborative environment, with opportunities to work alongside experienced researchers and gain hands-on experience in real-world applications. DO NOT miss this chance to grow your skills and make an impact in the field! Contact us today to learn more about available positions.

🏅 Awards

• Selected for CIPS Doctoral Dissertation Incentive Program, 2024（中国中文信息学会博士学位论文激励计划，Top 10 in China）
• Tsinghua University Excellent Doctoral Dissertation, 2024（清华大学优秀博士论文）
• ACL (CCF-A) Area Chair Award, 2023
• Microsoft Research Asia Fellowship Nomination Award, 2022 (Top 33 in Asia)
• National scholarship for doctoral students, 2022
• Excellent Graduate in Beijing, 2019
• Outstanding Graduate in Tsinghua University, 2019

Main Publications and Preprints

* indicates equal contribution; $\dagger$ indicates corresponding author(s)

A. Natural Language Generation

1. Zhuocheng Gong*, Ang Lv*, Jian Guan$^\dagger$, Junxi Yan, Wei Wu, Huishuai Zhang, Minlie Huang, Dongyan Zhao, Rui Yan. Mixture-of-Modules: Reinventing Transformers as Dynamic Assemblies of Modules. EMNLP 2024. [pdf]
2. Jian Guan, Jesse Dodge, David Wadden, Minlie Huang, Hao Peng. Language Models Hallucinate, but May Excel at Fact Verification. NAACL 2024. [pdf] [code]
3. Jiaxin Wen, Hao Zhou, Jian Guan, Jie Zhou, Minlie Huang. Re3Dial: Retrieve, Reorganize and Rescale Conversations for Long-Turn Open-Domain Dialogue Pre-training. EMNLP 2023. [pdf]
4. Jian Guan, Minlie Huang. Mitigating the Learning Bias towards Repetition by Self-Contrastive Training for Open-Ended Generation. Findings of ACL 2023 (Short Paper). [pdf] [code]
5. Xuekai Zhu*, Jian Guan*, Minlie Huang, Juan Liu. StoryTrans: Non-Parallel Story Author-Style Transfer with Discourse Representations and Content Enhancing. ACL 2023. [Area Chair Award] 🏅
6. Zhexin Zhang, Jian Guan, Xin Cui, Yu Ran, Bo Liu and Minlie Huang. Self-Supervised Sentence Polishing by Adding Engaging Modifiers. ACL 2023 (Demo).
7. Jian Guan, Zhenyu Yang, Rongsheng Zhang, Zhipeng Hu, Minlie Huang. Generating Coherent Narratives by Learning Dynamic and Discrete Entity States with a Contrastive Framework. AAAI 2023. [pdf] [code]
8. Zhexin Zhang*, Jian Guan*, Guowei Xu, Yixiang Tian, Minlie Huang. Automatic Comment Generation for Chinese Student Narrative Essays. EMNLP 2022 (Demo). [pdf] [code]
9. Jian Guan, Ziqi Liu, Minlie Huang. A Corpus for Understanding and Generating Moral Stories. NAACL 2022. [pdf] [data]
10. Jiaxin Wen, Zhexin Zhang, Jian Guan, Minlie Huang. Persona-Guided Planning for Controlling the Protagonist’s Persona in Story Generation. NAACL 2022. [pdf] [code]
11. Jian Guan, Zhuoer Feng, Yamei Chen, Ruilin He, Xiaoxi Mao, Changjie Fan, Minlie Huang. LOT: A Story-Centric Benchmark for Evaluating Chinese Long Text Understanding and Generation. TACL 2022. [pdf] [code]
12. Jian Guan, Zhexin Zhang, Zhuoer Feng, Zitao Liu, Wenbiao Ding, Xiaoxi Mao, Changjie Fan, Minlie Huang. OpenMEVA: A Benchmark for Evaluating Open-ended Story Generation Metrics. ACL 2021. [pdf] [toolkit]
13. Jian Guan, Xiaoxi Mao, changjie fan, Zitao Liu, Wenbiao Ding, Minlie Huang. Long Text Generation by Modeling Sentence-Level and Discourse-Level Coherence. ACL 2021. [pdf] [code]
14. Zhengyan Zhang, Yuxian Gu, Xu Han, Shengqi Chen, Chaojun Xiao, Zhenbo Sun, Yuan Yao, Fanchao Qi, Jian Guan, et al. CPM-2: Large-scale Cost-efficient Pre-trained Language Models. AI Open 2021. [pdf] [code]
15. Zhengyan Zhang, Xu Han, Hao Zhou, Pei Ke, Yuxian Gu, Deming Ye, Yujia Qin, YuSheng Su, Haozhe Ji, Jian Guan, et al. CPM: A Large-scale Generative Chinese Pre-trained Language Model. AI Open 2021. [pdf] [code]
16. Xiangzhe Kong, Jialiang Huang, Ziquan Tung, Jian Guan and Minlie Huang. Stylized Story Generation with Style-Guided Planning. Findings of ACL 2021. [pdf] [code]
17. Jian Guan, Minlie Huang. UNION: An Unreferenced Metric for Evaluating Open-ended Story Generation. EMNLP 2020. [pdf] [code] [bib]
18. Jian Guan, Fei Huang, Zhihao Zhao, Xiaoyan Zhu, Minlie Huang. A Knowledge-Enhanced Pretraining Model for Commonsense Story Generation. TACL 2020. [pdf] [code] [demo] [bib]
19. Jian Guan*, Yansen Wang*, Minlie Huang. Story ending generation with incremental encoding and commonsense knowledge. AAAI 2019. [pdf] [code] [bib]
20. Pei Ke, Jian Guan, Minlie Huang, Xiaoyan Zhu. Generating informative responses with controlled sentence function. ACL 2018. [pdf] [code] [bib]
21. Jianzhu Yao, Ziqi Liu, Jian Guan, Minlie Huang. A Benchmark for Understanding and Generating Dialogue between Characters in Stories. arxiv preprint 2022. [pdf]
22. Yuan Yao, Qingxiu Dong, Jian Guan, et al. CUGE: A chinese language understanding and generation evaluation benchmark. arxiv preprint 2021. [pdf]
23. Fei Huang, Jian Guan, Pei Ke, Qihan Guo, Xiaoyan Zhu, Minlie Huang. A Text GAN for Language Generation with Non-Autoregressive Generator. arxiv preprint 202. [pdf]
24. Fei Huang, Dazhen Wan, Zhihong Shao, Pei Ke, Jian Guan, Yilin Niu, Xiaoyan Zhu, Minlie Huang. CoTK: An Open-Source Toolkit for Fast Development and Fair Evaluation of Text Generation. arxiv preprint 2020. [pdf] [code]

B. Advanced Reasoning in LLMs and LVLMs

1. Chuanqi Cheng*, Jian Guan*, Wei Wu, Rui Yan. Scaling Video-Language Models to 10K Frames via Hierarchical Differential Distillation. ICML 2025 [pdf] [code] [7B SOTA LVLM on [VideoMME](https://video-mme.github.io/home_page.html)]
2. Xueliang Zhao, Wei Wu, Jian Guan, Lingpeng Kong. Promptcot: Synthesizing olympiad-level problems for mathematical reasoning in large language models. Findings of ACL 2025. [pdf] [code]
3. Jiaxin Wen*, Jian Guan*, Hongning Wang, Wei Wu, Minlie Huang. Unlocking Reasoning Potential in Large Langauge Models by Scaling Code-form Planning. ICLR 2025. [pdf] [data]
4. Jia-Nan Li*, Jian Guan*, Wei Wu, Zhengtao Yu, Rui Yan. 2D-TPE: Two-Dimensional Positional Encoding Enhances Table Understanding for Large Language Models. WWW 2025. [pdf]
5. Jian Guan, Wei Wu, Zujie Wen, Peng Xu, Hongning Wang, Minlie Huang. AMOR: A Recipe for Building Adaptable Modular Knowledge Agents Through Process Feedback. NeurIPS 2024. [pdf]
6. Chuanqi Cheng*, Jian Guan*, Wei Wu, Rui Yan. From the Least to the Most: Building a Plug-and-Play Visual Reasoner via Data Synthesis. EMNLP 2024. [pdf]

C. Personalized Alignment

1. Jia-Nan Li*, Jian Guan*, Wei Wu, Rui Yan. Extended Inductive Reasoning for Personalized Preference Inference from Behavioral Signals. arxiv preprint 2025. [pdf] [code]
2. Jia-Nan Li*, Jian Guan*, Songhao Wu, Wei Wu, Rui Yan. From 1,000,000 users to every user: Scaling up personalized preference for user-level alignment. arxiv preprint 2025. [pdf] [data]
3. Jian Guan, Junfei Wu, Jia-Nan Li, Chuanqi Cheng, Wei Wu. A Survey on Personalized Alignment–The Missing Piece for Large Language Models in Real-World Applications. Findings of ACL 2025. [pdf]
4. Zhuocheng Gong, Jian Guan, Wei Wu, Huishuai Zhang, Dongyan Zhao. Latent Preference Coding: Aligning Large Language Models via Discrete Latent Codes. ICML 2025 [pdf]

D. Others

1. Guhao Feng, Yihan Geng, Jian Guan, Wei Wu, Liwei Wang, Di He. Theoretical Benefit and Limitation of Diffusion Language Model. arxiv preprint 2025. [pdf]
2. Sahand Sabour et al. Human Decision-making is Susceptible to AI-driven Manipulation. arxiv preprint 2025. [pdf]
3. Xueying Bai*, Jian Guan*, Hongning Wang. A Model-Based Reinforcement Learning with Adversarial Training for Online Recommendation. NeurIPS 2019. [pdf] [code] [bib]

Teaching

He was a TA for the following undergraduate courses:

• Artificial Neural Network (2019 Fall, 2020 Fall, 2021 Fall)
• Object-Oriented Programming (2019 Spring, 2020 Spring, 2021 Spring, 2022 Spring)