Zhihao Xu - Renmin University of China

I am a master's student (2024–2027) at the Gaoling School of Artificial Intelligence, Renmin University of China. My research interests include LLM Alignment & RLHF, LLM Agentic Training. I have published 6 papers as the first/co-first author in NeruIPS, ICLR, ICML and ACL.

I expect to graduate in June 2027 and am actively seeking opportunities for intern positions (foundation model team, agentic training). Please feel free to reach out.

Selected Publications

AORCHESTRA: Automating Sub-Agent Creation for Agentic Orchestration
Jianhao Ruan*, Zhihao Xu*, Yiran Peng, Fashen Ren, Zhaoyang Yu, Xinbing Liang, Jinyu Xiang, Yongru Chen, Bang Liu, Chenglin Wu, Yuyu Luo, Jiayi Zhang
ICML 2026, 100+ Github Stars

Unlocking Implicit Experience: Synthesizing Tool-Use Trajectories from Text
Zhihao Xu, Rumei Li, Jiahuan Li, Rongxiang Weng, Jingang Wang, Xunliang Cai, Xiting Wang
ACL 2026 Main Conference

PAM: Enhancing General Alignment of Large Reasoning Models through Priority-Aware Metacognition
Zhihao Xu, Fuzhen Yang, Liang Lin, Xiting Wang
ACL 2026 Main Conference

Reward Consistency: Improving Multi-Objective Alignment from a Data-Centric Perspective
Zhihao Xu, Yongqi Tong, Xin Zhang, Jun Zhou, Xiting Wang
ACL 2026 Findings

OrthAlign: Orthogonal Subspace Multi-Objective Alignment
Liang Lin, Zhihao Xu*, Junhao Dong, Jian Zhao, Yuchen Yuan, Guibin Zhang, Miao Yu, Yiming Zhang, Zhengtao Yao, Huahui Yi, Dongrui Liu, Xinfeng Li, Kun Wang
ICLR 2026 [Paper]

Uncovering Safety Risks of Large Language Models through Concept Activation Vector
Zhihao Xu, Ruixuan Huang, Changyu Chen, Xiting Wang
NeurIPS 2024 · 80+ Citations [Paper]

Internal Value Alignment in Large Language Models through Controlled Value Vector Activation
Haoran Jin*, Meng Li*, Xiting Wang, Zhihao Xu, Minlie Huang, Yantao Jia, Defu Lian
ACL 2025 Main Conference [Paper]

Evaluating Readability and Faithfulness of Concept-based Explanations
Meng Li*, Haoran Jin*, Ruixuan Huang, Zhihao Xu, Defu Lian, Zijia Lin, Di Zhang, Xiting Wang
EMNLP 2024 [Paper]

Experience

Intern · Alibaba Qwen Team
2025.04 - 至今
Work on skills/search agent.

Research Intern · Meituan LongCat Team(美团)
2025.11-2026.04
Work on agentic training.

Research Intern · Ant Group (蚂蚁集团)
2024.08-2025.05
Work on LLM RLHF Post Training.

Research Intern · Microsoft Research Asia (MSRA)
Research on LLM alignment.

Contributor · MetaGPT/OpenManus Team
Work on agentic orchestration.

Zhihao Xu (许志豪)

About Me

Selected Publications

Experience