I am currently a postdoctoral researcher at the NLP Laboratory of Fudan University, working under the supervision of Prof. Xuanjing Huang. My postdoctoral tenure is expected to conclude in September 2025. Prior to joining Fudan NLP Lab, I got both my B.S. (2017) and Ph.D. (2023) degree in Computer Science from East China Normal University, advised by Prof. Yuanbin Wu and Xiaoling Wang.

Research

My primary research interests include the architectural design of LLMs, with a particular focus on ➊Efficient infrastructure via algorithm-hardware co-design, ➋Language-centric multimodal in-context learning, and ➌Native long-context extrapolation in LLMs.

May 2025

🎉Congratulations to Yufang Liu on earning her Ph.D.! I'm honored to have co-advised her with Prof. Yuanbin Wu on her AAAI21, EMNLP24, and ACL25 work. I deeply appreciate her self-driven, focused, and proactive approach to research and her well-rounded abilities. She will soon join Meituan Research. Best wishes.

🎉Congratulations to Yanting Liu on completing her Master's degree! I was delighted to co-advise her on her EMNLP24 work. She is intelligent and hardworking. While I feel a bit regretful that she chose not to pursue a Ph.D., I’m glad to see her moving forward—joining ByteDance to work on multimodal models. Best wishes.

* indicates equal contribution, # indicates corresponding author

ACL25

Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMs

Tao Ji, Bin Guo, Yuanbin Wu, Qipeng Guo, Lixing Shen, Zhan Chen, Xipeng Qiu, Qi Zhang, Tao Gui#

ACL25

The Role of Visual Modality in Multimodal Mathematical Reasoning: Challenges and Insights

Yufang Liu*, Yao Du*, Tao Ji#, Jianing Wang, Yang Liu, Yuanbin Wu#, Aimin Zhou#, Mengdi Zhang, Xunliang Cai

EMNLP24

Investigating and Mitigating Object Hallucinations in Pretrained Vision-Language (CLIP) Models

Yufang Liu*, Tao Ji*, Changzhi Sun, Yuanbin Wu, Aimin Zhou

EMNLP24
Findings

LongHeads: Multi-Head Attention is Secretly a Long Context Processor

Yi Lu, Xin Zhou, Wei He, Jun Zhao, Tao Ji#, Tao Gui#, Qi Zhang#, Xuanjing Huang

ACL24
Findings

Length Generalization of Causal Transformers without Position Encoding

Jie Wang*, Tao Ji*, Yuanbin Wu#, Hang Yan, Tao Gui, Qi Zhang, Xuanjing Huang, Xiaoling Wang#

ACL19

Graph-based Dependency Parsing with Graph Neural Networks

Tao Ji, Yuanbin Wu, Man Lan