Jiazhan (Jamie) Feng
fengjiazhan <at> pku.edu.cn OR im.jzfeng <at> gmail.com
"Jiazhan" is pronounced as <drazen>. I go by "Jamie". 🎸
Embedded Files
Hi! I am Jiazhan (Jamie) Feng, a Ph.D. student at the School of Intelligence Science and Technology, Peking University, advised by Prof. Dongyan Zhao and Prof. Rui Yan. Currently, I am a Research Scientist in ByteDance Seed, and also a Visiting (Recognised) Student at the University of Oxford, advised by Prof. Yee Whye Teh. Before that, I was a research intern at Microsoft Azure AI, Redmond, advised by Dr. Ruochen Xu, Dr. Yelong Shen and Dr. Weizhu Chen, and a research intern at Microsoft Research Asia (MSRA).
Hi! I am Jiazhan (Jamie) Feng, a Ph.D. student at the School of Intelligence Science and Technology, Peking University, advised by Prof. Dongyan Zhao and Prof. Rui Yan. Currently, I am a Research Scientist in ByteDance Seed, and also a Visiting (Recognised) Student at the University of Oxford, advised by Prof. Yee Whye Teh. Before that, I was a research intern at Microsoft Azure AI, Redmond, advised by Dr. Ruochen Xu, Dr. Yelong Shen and Dr. Weizhu Chen, and a research intern at Microsoft Research Asia (MSRA).
🚀 Feel free to email me for any form of academic cooperation. Cheers! ^_^
Research Interest 💡
Research Interest 💡
I am fascinated by real-world problems related to multimodal agents, and machine learning.
My recent publications conducted specific research on the following problems:
Agentic-LLMs: ReTool-RL for Strategic Tool Use (Arxiv '25). UI-TARS-2-a native GUI agent trained with multi-turn agent RL (Arxiv '25)
LLM Reasoning: WizardLM-Empowering LLMs to follow complex instructions (ICLR '24), LoGiPT-Logical reasoning with LLMs (NAACL '24), InteR-LLMs for IR (ACL '24).
News 🗞️
News 🗞️
[Sep 2025]: We release UI-TARS-2 technical report 🚀.
[Apr 2025]: Our ReTool paper and recipe is available and I will attend ICLR '25 in Singapore.
[Oct 2024]: Our paper got accepted to WSDM 2025. 🎉
[May 2024]: Our paper got accepted to ACL 2024. 🎉 Looking forward to seeing everyone in Bangkok!
[Mar 2024]: Our paper got accepted to NAACL 2024. 🎉 Looking forward to seeing everyone in Mexico City!
[Mar 2024]: Our WizardLM got accepted to ICLR 2024. 🎉
News before Mar 2024 is not included in this list.
2025
2025
UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning.
UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning.
ByteDance Seed (Core Contributor).
In ArXiv 2509.02544. [Technical Report] [Demo] [new]
AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting.
AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting.
Shijue Huang, Hongru Wang, Wanjun Zhong, Zhaochen Su, Jiazhan Feng, Bowen Cao and Yi R. Fung.
In ArXiv 2505.18822. [Arxiv version] [new]
ReTool: Reinforcement Learning for Strategic Tool Use in LLMs.
ReTool: Reinforcement Learning for Strategic Tool Use in LLMs.
Jiazhan Feng^, Shijue Huang^, Xingwei Qu, Ge Zhang, Yujia Qin, Baoquan Zhong, Chengquan Jiang, Jinxin Chi and Wanjun Zhong.
In Arxiv 2504.11536. [Arxiv version] [Project Page] [Training Recipe] [new]
Seed-Thinking-v1.5: Advancing Superb Reasoning Models with Reinforcement Learning.
Seed-Thinking-v1.5: Advancing Superb Reasoning Models with Reinforcement Learning.
ByteDance Seed.
In Arxiv 2504.13914. [Technical Report] [Project Page]
RetriEVAL: Evaluating Text Generation with Contextualized Lexical Match.
RetriEVAL: Evaluating Text Generation with Contextualized Lexical Match.
Zhen Li*, Xinchi Li, Chongyang Tao, Jiazhan Feng, Tao Shen, Can Xu, Hao Wang, Dongyan Zhao and Shuai Ma.
In The Eighteenth International Conference on Web Search and Data Mining (WSDM2025). [paper]
2024
2024
Synergistic Interplay between Search and Large Language Models for Information Retrieval.
Synergistic Interplay between Search and Large Language Models for Information Retrieval.
Jiazhan Feng, Chongyang Tao, Xiubo Geng, Tao Shen, Can Xu, Guodong Long, Dongyan Zhao and Daxin Jiang.
In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024). [paper] [code]
Language Models can be Deductive Solvers.
Language Models can be Deductive Solvers.
Jiazhan Feng, Ruochen Xu, Junheng Hao, Hiteshi Sharma, Yelong Shen, Dongyan Zhao and Weizhu Chen.
In Findings of the Association for Computational Linguistics: NAACL 2024 (NAACL 2024 Findings). [paper] [project page]
WizardLM: Empowering Large Pre-Trained Language Models to Follow Complex Instructions.
WizardLM: Empowering Large Pre-Trained Language Models to Follow Complex Instructions.
Can Xu, Qingfeng Sun, Kai Zheng, Xiubo Geng, Pu Zhao, Jiazhan Feng, Chongyang Tao, Qingwei Lin and Daxin Jiang.
In Twelfth International Conference on Learning Representations (ICLR 2024). [paper] [code and demo]
2023
2023
Length-Adaptive Distillation: Customizing Small Language Model for Dynamic Token Pruning.
Length-Adaptive Distillation: Customizing Small Language Model for Dynamic Token Pruning.
Chang Liu, Chongyang Tao, Jianxin Liang, Jiazhan Feng, Tao Shen, Quzhe Huang and Dongyan Zhao.
In Findings of the Association for Computational Linguistics: EMNLP 2023 (EMNLP 2023 Findings). [paper]
MMDialog: A Large-scale Multi-turn Dialogue Dataset Towards Multi-modal Open-domain Conversation.
MMDialog: A Large-scale Multi-turn Dialogue Dataset Towards Multi-modal Open-domain Conversation.
Jiazhan Feng, Qingfeng Sun, Can Xu, Pu Zhao, Yaming Yang, Chongyang Tao, Dongyan Zhao and Qingwei Lin.
In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL 2023). [paper] [project page]
CORE: Cooperative Training of Retriever-Reranker for Effective Dialogue Response Selection.
CORE: Cooperative Training of Retriever-Reranker for Effective Dialogue Response Selection.
Chongyang Tao, Jiazhan Feng, Tao Shen, Chang Liu, Juntao Li, Xiubo Geng and Daxin Jiang.
In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL 2023). [paper]
FAA: Fine-grained Attention Alignment for Cascade Document Ranking.
FAA: Fine-grained Attention Alignment for Cascade Document Ranking.
Zhen Li*, Chongyang Tao, Jiazhan Feng, Tao Shen, Dongyan Zhao, Xiubo Geng and Daxin Jiang.
In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL 2023). [paper]
Attend, Select and Eliminate: Accelerating Multi-turn Response Selection with Dual-attention-based Content Elimination.
Attend, Select and Eliminate: Accelerating Multi-turn Response Selection with Dual-attention-based Content Elimination.
Jianxin Liang, Chang Liu, Chongyang Tao, Jiazhan Feng and Dongyan Zhao.
In Findings of the Association for Computational Linguistics: ACL 2023 (ACL 2023 Findings). [paper]
Dimension-Prompts Boost Commonsense Consolidation.
Dimension-Prompts Boost Commonsense Consolidation.
Jiazhan Feng, Chongyang Tao, Tao Shen, Chang Liu and Dongyan Zhao.
In 46th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2023). [paper]
Learning Multi-turn Response Selection in Grounded Dialogues with Reinforced Knowledge and Context Distillation.
Learning Multi-turn Response Selection in Grounded Dialogues with Reinforced Knowledge and Context Distillation.
Jiazhan Feng, Chongyang Tao, Xueliang Zhao and Dongyan Zhao.
In ACM Transactions on Information Systems, Volume 41, Issue 4 (TOIS, CCF Rank A). [paper]
2022
2022
How to Represent Context Better? An Empirical Study on Context Modeling for Multi-turn Response Selection.
How to Represent Context Better? An Empirical Study on Context Modeling for Multi-turn Response Selection.
Jiazhan Feng, Chongyang Tao, Chang Liu, Rui Yan and Dongyan Zhao.
In Findings of the Association for Computational Linguistics: EMNLP 2022 (EMNLP 2022 Findings). [paper]
Rethinking Task-Specific Knowledge Distillation: Contextualized Corpus as Better Textbook.
Rethinking Task-Specific Knowledge Distillation: Contextualized Corpus as Better Textbook.
Chang Liu, Chongyang Tao, Jianxin Liang, Tao Shen, Jiazhan Feng, Quzhe Huang and Dongyan Zhao.
In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP 2022). [paper]
Reciprocal Learning of Knowledge Retriever and Response Ranker for Knowledge-Grounded Conversations.
Reciprocal Learning of Knowledge Retriever and Response Ranker for Knowledge-Grounded Conversations.
Jiazhan Feng, Chongyang Tao, Zhen Li*, Chang Liu, Tao Shen and Dongyan Zhao.
In Proceedings of the 29th International Conference on Computational Linguistics (COLING 2022). [paper]
Training Two-Stage Knowledge-grounded Dialogues with Attention Feedback.
Training Two-Stage Knowledge-grounded Dialogues with Attention Feedback.
Zhen Li*, Jiazhan Feng, Chongyang Tao and Dongyan Zhao.
In 11th CCF International Conference on Natural Language Processing and Chinese Computing (NLPCC 2022). [paper]
Multi-Granularity Structural Knowledge Distillation for Language Model Compression.
Multi-Granularity Structural Knowledge Distillation for Language Model Compression.
Chang Liu, Chongyang Tao, Jiazhan Feng and Dongyan Zhao.
In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (ACL 2022). [paper]
2021
2021
A Pre-training Strategy for Zero-Resource Response Selection in Knowledge-Grounded Conversations.
A Pre-training Strategy for Zero-Resource Response Selection in Knowledge-Grounded Conversations.
Chongyang Tao, Changyu Chen, Jiazhan Feng, Ji-Rong Wen and Rui Yan.
In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL 2021). [paper]
Response Ranking with Multi-Types of Deep Interactive Representations in Retrieval-based Dialogues.
Response Ranking with Multi-Types of Deep Interactive Representations in Retrieval-based Dialogues.
Ruijian Xu, Chongyang Tao, Jiazhan Feng, Wei Wu, Rui Yan and Dongyan Zhao.
In ACM Transactions on Information Systems, Volume 39, Issue 4 (TOIS, CCF Rank A). [paper]
A Survey on Response Selection for Retrieval-based Dialogues.
A Survey on Response Selection for Retrieval-based Dialogues.
Chongyang Tao, Jiazhan Feng, Rui Yan, Wei Wu and Daxin Jiang.
In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence (IJCAI 2021). [paper]
2019
2019
Learning a Matching Model with Co-teaching for Multi-turn Response Selection in Retrieval-based Dialogue Systems.
Learning a Matching Model with Co-teaching for Multi-turn Response Selection in Retrieval-based Dialogue Systems.
Jiazhan Feng, Chongyang Tao, Wei Wu, Yansong Feng, Dongyan Zhao and Rui Yan.
In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL 2019). [paper]
Preprints 📎
Preprints 📎
A Step Closer to Comprehensive Answers: Constrained Multi-Stage Question Decomposition with Large Language Models.
A Step Closer to Comprehensive Answers: Constrained Multi-Stage Question Decomposition with Large Language Models.
Hejing Cao, Zhenwei An, Jiazhan Feng, Kun Xu, Liwei Chen and Dongyan Zhao.
Teaching Text-to-Image Models to Communicate in Dialog.
Teaching Text-to-Image Models to Communicate in Dialog.
Xiaowen Sun^*, Jiazhan Feng^, Yuxuan Wang, Yuxuan Lai, Xingyu Shen and Dongyan Zhao.
Professional Services 🎓
Professional Services 🎓
Area Chairs (AC): ACL'25, EMNLP'25
Program Committee (Reviewer): NeurIPS'24-, ICML'25, ACL'21-, EMNLP'22-, NAACL'22-, COLING'24-, ACL Rolling Review.
Page updated
Google Sites
Report abuse