Zijian Wang

Zijian Wang

Incoming Ph.D. Student at Shanghai Jiao Tong University

Hi there👋, I'm Zijian Wang. I'm mainly interested in Reinforcement Learning and LLM Agents and have conducted a series of studies on Agentic RL and LLM Agents System.

Currently, I am doing a research internship at X-LANCE Lab at Shanghai Jiao Tong University and am responsible for the Agent Memory & LLM4Research within the LLM group, supervised by Prof. Lu Chen.

From 2026 Fall, I will be a Ph.D. Student at X-LANCE Lab, Shanghai Jiao Tong University.

Selected Publications

Walk Wisely on Graph: Knowledge Graph Reasoning with Dual Agents via Efficient Guidance-Exploration

Zijian Wang, Bin Wang, Haifeng Jing, Huayu Li, Hongbo Dou

AAAI Conference on Artificial Intelligence (AAAI), 2025

A dual-agent framework for knowledge graph reasoning, which combines an efficient guidance agent with an exploration agent to achieve better performance with less computational resources.

Knowledge Graph Reinforcement Learning Reasoning
Paper Preview Paper Preview

Windows deep transformer Q-networks: an extended variance reduction architecture for partially observable reinforcement learning

Zijian Wang, Bin Wang, Hongbo Dou, Zhongyuan Liu

Applied Intelligence, 2024

A transformer-based architecture for dealing with partially observable reinforcement learning environments by reducing variance through windowed state processing.

Reinforcement Learning Transformer POMDP
Paper Preview Paper Preview

Externalizing Research Synthesis and Validation in AI Scientists through a Research Harness

Zijian Wang*, Hanqi Li*, Ziyue Yang, Zijian Hu, Shenghan Zuo, Yunzhe Zhang, Da Ma, Danyu Luo, Chenrun Wang, Jing Peng, Tiancheng Huang, Sijia Guo, Huayang Wang, Zichen Zhu, Senyu Han, Yilu Cao, Kai Yu, Lu Chen

arXiv preprint arXiv:2606.18874

A research harness that externalizes research synthesis and experimental validation into explicit, inspectable, and controllable structures for AI scientists

AI4Research LLM Agents System Recursive Self-Improvement
Paper Preview Paper Preview

AirQA: A Comprehensive QA Dataset for AI Research with Instance-Level Evaluation

Tiancheng Huang, Ruisheng Cao, Yuxin Zhang, Zhangyi Kang, Zijian Wang, Chenrun Wang, Yijie Luo, Hang Zheng, Lirong Qian, Lu Chen, Kai Yu

International Conference on Learning Representations (ICLR), 2026

A human-annotated comprehensive scientific paper QA dataset with 13948 papers and 1246 questions, that encompasses multi-task, multi-modal and instance-level evaluation.

Large Language Model Mutli-hop QA Dataset and Benchmark
Paper Preview Paper Preview

G-Scaler: Graph Few-Shot Class-Incremental Learning via Structural Coupling and Elegant Shift Rectification

Bin Wang, Boxiang Tao, Hongbo Dou, Zijian Wang, Lin Luo

Neurocomputing, 2026

A graph learning method that enables few-shot class-incremental learning through structural coupling mechanisms and shift rectification strategies.

Graph Learning Few-Shot Learning Incremental Learning
Paper Preview Paper Preview

Honors & Service 🏆

Selected Awards

Major Honors

National Scholarship

Ministry of Education (Top 0.2%)

2024

First Prize

The Chinese Mathematics Competitions for College Students

2023

Outstanding Student Prize

China University of Petroleum

2023

2024

Outstanding Star Scholarship for Innovation and Entrepreneurship

China University of Petroleum

2024

2025

Community Service

Journal / Conference Reviewer

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)

AAAI Conference on Artificial Intelligence (AAAI)

ACL Rolling Review (ARR) · Association for Computational Linguistics

Hobbies & Interests 🎵

Music

Music

Music has always been an essential part of my life. I enjoy various genres, but Japanese pop music and anisong compositions hold a special place in my heart. My favorite songs include NEO SKY NEO MAP!, ユニゾン, and 満ちてゆく, whose melodies accompany me during research and study sessions.

Idol Groups

Anime Idol Groups

I'm an enthusiastic fan of anime idol groups, particularly drawn to their choreography, music production, and the positive energy they bring. My favorite group is Poppin' Party, 虹ヶ咲学園スクールアイドル同好会, and スタァライト九九組, whose self-producing capabilities and diverse performances have always impressed me.

Following their journey not only brings joy but also inspires me to pursue excellence in my own work. I admire the dedication and teamwork displayed by these performers, which often reminds me of the collaborative nature of research. One of my current dreams is to go and watch their live performances on the spot! 🥰

Outdoor Travel

Outdoor Travel

Exploring natural landscapes and diverse cultures has become my favorite way to recharge. I've been fortunate to visit several breathtaking locations, including Zhu Shan in Qingdao, Jinsha Jiang in Panzhihua, and the 3rd Beach in Qingdao.

Each journey provides new perspectives and inspiration for my research. The tranquility found in nature often helps me develop fresh ideas and approaches to academic challenges. Hiking through mountain trails has taught me perseverance, while observing different ecosystems has broadened my understanding of complex systems.