/   /   /   / 

Hello, World!

I am Weihao Liu, currently a research intern at Microsoft (Beijing), mentored by Wenbiao Ding and Dr. Ming Gong. Previously, I was a research assistant at Institute of Automation, Chinese Academy of Sciences, advised by Prof. Kang Liu, where I started my journey in NLP. I obtained my bachelor's degree in Computer Science from University of Science and Technology Beijing (USTB).

My research interest lies in natural language processing, particularly in building versatile, trustworthy and interpretable large language model (LLM) systems. My recent research mainly includes the following aspects:

• Long-context Modeling: I am now working on improving the long-context modeling abilities of LLMs, especially exploring the effectiveness of attention mechanisms when LLMs meet extremely long-form inputs.
• Trustworthy NLP: Exploring methods to reduce hallucinations in LLMs and improve the truthfulness of their responses. This includes investigating both prompt engineering approaches (e.g., CoT, RAG) and model optimization techniques (e.g., RLHF, DPO).
• Understanding and Reasoning across modalities and languages: Exploring how LLMs can process inputs and knowledge from different modalities and languages, particularly in tasks such as visual-text reasoning, table-text reasoning and multilingual reasoning.

I am currently looking for Phd positions for Fall 2025.

• [2025.02] Our work MuDAF is available on arXiv!
• [2024.12] One paper accepted by COLING 2025 as an ORAL presentation!

• Silver Medal in ACM-ICPC Asia-East Continent Final, 2022
• National Scholarship, China, 2021
• Bronze Medal in National Olympiad in Informatics (NOI), China, 2019

• Teaching Assistant of Programming Fundamentals A, USTB, SCCE, 2021, 2022