Haiquan Zhao’s HomePage

About me.

I am an Algorithm Engineer on the Qwen Team at Alibaba Group, where I focus on Large Language Model (LLM) safety and decoding strategies. I work closely with Dr. Fei Huang.

Prior to joining Alibaba, I earned my Master’s degree from Fudan University under the supervision of Prof. Zhixu Li. During this time, I collaborated closely with Dr. Xuwu Wang, PhD student Shisong Chen, and independent researcher Jiaan Wang. My research interests lie primarily in natural language processing and large language models, with a particular emphasis on LLM safety and emotion-aware agents. I have also conducted research in the area of Multimodal Knowledge Graphs (MMKG).

I received my Bachelor’s degree in Computer Science and Technology from Wuhan University.

News:

• OVEL OVEL: Online Video Entity Linking was accpeted by COLING 2025!

• ESC-Eval ESC-Eval: Evaluating Emotion Support Conversations in Large Language Models was accpeted by EMNLP 2024 (Oral)!

Education:

• B.S. in Wuhan, Wuhan University, 2022.
• M.S. in Shanghai, Fudan University, 2025 (expected).

Internships

1. Research Intern on Safety of Large Language Model. (Qwen Team 2024.09-)

2. Algorithm Intern on Large Language Model data engineering. (Tencent 2024.6-2024.09)

3. Research Intern on Emotional Intelligence. (Shanghai AI lab 2023.11-2024.06)

4. Network asset identification based on natural language processing feature engineering. (Huashun Xinan 2022.6-2022.8)

Publications.

• Kaichen Zhao, Yaoxian Song, Haiquan Zhao, Haoyu Liu, Tiefeng Li, Zhixu Li: Towards Coarse-grained Visual Language Navigation Task Planning Enhanced by Event Knowledge Graph. CIKM 2024.

• Haiquan Zhao, Lingyu Li, Shisong Chen, Shuqi Kong, Jiaan Wang, Kexin Huang, Tianle Gu, Yixu Wang, Dandan Liang, Zhixu Li, Yan Teng, Yanghua Xiao, Yingchun Wang: ESC-Eval: Evaluating Emotion Support Conversations in Large Language Models. EMNLP 2024.

• Haiquan Zhao, Xuwu Wang, Shisong Chen, Zhixu Li, Xin Zheng, Yanghua Xiao: OVEL: Large Language Model as Memory Manager for Online Video Entity Linking. COLING 2025.

• 赵海全,王续武,李金亮,李直旭,肖仰华.面向视频的细粒度多模态实体链接.软件学报,2023

• A network asset identification method based on natural language processing feature engineering. (Patent:CN115733903A)

Projects.

1. Long document Keypoint extraction based on chatpdf. (Xingdian 2023.11-2024.07)

2. Semi-automatic schema construction based on course introduction. (Huawei 2022.11-2023.10)

3. Fine-grained multimodal entity linking for video. (Fudan University 2022.12-2023.7)

Research Activities:

Reviewer: ARR 2025.

Others:

• Backend management project front-end page;
• Reservoir flow query front-end page;
• Alumni second-hand website back-end query page;
• Nlp-begginner (text classification, sequence Annotation, language model and so on.);
• network protocol text named entity recognition, entity relationship extraction.