About Me

I am currently a 2nd-year Ph.D. student at the University of Science and Technology of China (USTC), fortunate to be co-advised by Prof. Zheng-Jun Zha and Associate Prof. Jiawei Liu. Previously, I received my Bachelor’s degree in Automation from Northeastern University (2019-2023).

Research Interests:

• Multimodal Reasoning and Understanding - Developing models capable of understanding and reasoning over complex multimodal and long-horizon information
• Self-Evolving Agents - Enabling agents to continuously acquire, refine, and internalize reusable knowledge and strategies through interaction and experience
• Agentic Reinforcement Learning — Training general / domain-specific agents with reinforcement learning for stronger reasoning, tool use, and long-horizon decision-making

🔥 News

📖 Education

📝 Publications

Mining the Potential of Rehearsal Mechanism for VLM-based Class Incremental Learning

Sen Tao, Jiawei Liu, Yongchao Xu, Guangxi Wan, Peng Zeng.
Chinese Conference on Computer Vision and Machine Intelligence (CVM), 2026.
PDF Code

We propose a debiased memory-calibrated Gaussian discriminant analysis method for VLM-based class incremental learning.

Learning to Diversify and Focus: A Reinforcement Framework for Open-Vocabulary HOI Detection

Yongchao Xu, Jiawei Liu, Junfeng Wang, Sen Tao, Na Jiang, Zheng-Jun Zha.
Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR), 2026.
PDF Supp Code

We propose a semantic-diversified and interaction-focused framework for open-vocabulary HOI detection.

PSP: Prompt-Guided Self-Training Sampling Policy for Active Prompt Learning

Sen Tao, Kaiduo Feng, Jiawei Liu, Peng Zeng, Yongchao Xu, Yufei Zheng, Zheng-Jun Zha.
International Conference on Learning Representations (ICLR), 2026.
PDF Code

We propose a prompt-guided self-training sampling policy for active prompt learning.

Mamba-Driven Comprehensive Context Learning for Zero-Shot HOI Detection

Jiawei Liu, Yongchao Xu (co-first author), Sen Tao, Yuexuan Qi, Zheng-Jun Zha.
International Journal of Computer Vision (IJCV), 2026.
PDF Code

We propose a comprehensive context learning framework for zero-shot HOI detection.

Boosting Active Prompt Learning via Discriminative Self-Training Dual-Curriculum Learning

Sen Tao, Jiawei Liu, Peng Zeng, Yongchao Xu, Bingyu Hu, Zheng-Jun Zha.
International Journal of Computer Vision (IJCV), 2026.
PDF Code

We propose a dual-curriculum active prompt learning framework.

HOIMamba: Efficient Mamba-based Disentangled Progressive Learning for HOI Detection

Yongchao Xu, Jiawei Liu, Sen Tao, Qiang Zhang, Zheng-Jun Zha.
The Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI), 2025.
PDF Code

We introduce a Mamba-based decoder for robust HOI detection.

Hierarchical Knowledge Prompt Tuning for Multi-task Test-Time Adaptation

Qiang Zhang, Mengsheng Zhao, Jiawei Liu, Fanrui Zhang, Yongchao Xu, Zheng-Jun Zha.
Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR), 2025.
PDF Code

We address multi-task test-time adaptation for pre-trained VLMs.

ESCNet: Entity-enhanced and Stance Checking Network for Multimodal Fact-Checking

Fanrui Zhang, Jiawei Liu, Jingyi Xie, Qiang Zhang, Yongchao Xu, Zheng-Jun Zha.
International World Wide Web Conference (WWW), 2024.
PDF Code

We establish a large-scale, multi-domain Chinese multimodal fact-checking dataset.

🏆 Honors & Awards