Professor Mengyue WU is an Associate Professor and Ph.D. Supervisor at the School of Computer Science (School of Cybersecurity), Shanghai Jiao Tong University, and a member of the Cross-Media Language Intelligence Lab (X-LANCE). She holds a Bachelor's degree from Beijing Normal University and a Ph.D. from the University of Melbourne, Australia. Before her current role, she was recognized as a Shanghai Pujiang Talent. Her broad research interests include audio signal processing and natural language processing for computational mental health, with a specific focus on empathy in large language models, proactive dialogue systems, and pathological audio analysis.

She has received numerous prestigious awards, including the 2025 ACL Outstanding Paper Award, the 2023 NCMMSC Best Paper Nomination Award, and was an International Champion in two tasks at the 2022 IEEE DCASE Challenge. Her work has also earned her the Second Prize at the Shanghai Digital Medical Technology and Application Innovation Competition for "Research and Application of Large Language Models for Mental Health." Her research is supported by the National Natural Science Foundation of China, where she has led both General and Youth Projects. She has published over 70 papers in leading conferences and journals such as ICLR, ACL, EMNLP, and IEEE/ACM T-ASLP.

She serves as an Area Chair for ARR Rolling Review and Interspeech. Within the academic community, she is the Deputy Secretary-General of the CCF Technical Committee on Affective Computing and an Executive Committee Member of the CCF Technical Committee on Speech and Dialogue. She has also been a member of IEEE and ACM since 2018. 

Her research is supported by collaborations with leading industry and medical institutions, including Ant Group, Fudan University's Institute of Ethics in Science and Technology, Peking Union Medical College Hospital, and Ruijin Hospital. This partnership led to the co-publication of the "White Paper on Ethics and Safety in Medical Large Language Models." Her work has resulted in several open-source platforms and tools, such as the "Mental Disorder Symptom Detection Model" and the "Ethical Safety Alignment Framework for Medical Large Language Models," providing crucial resources for enhancing AI empathy and security in enterprise and healthcare settings. Her full reserach profile can be accessed here Google Scholar

Research Interests

Her research broadly encompasses audio signal processing, natural language processing, and their application to computational mental health. Currently, her work focuses on building empathetic, secure, and reliable language models and applying them specifically to psychological counseling and pathological analysis. Some of her key research areas are highlighted below.

Digital Mental Health & Audio Processing

Her research in this area is motivated by the goal of using AI to understand and support mental well-being. A key focus is on developing proactive dialogue systems capable of empathetic interaction and identifying mental health indicators from conversation. This work extends to pathological audio analysis, where she investigates how vocal biomarkers can be used for detecting and monitoring mental health conditions. Her lab has developed models for detecting symptoms of mental disorders from both acoustic and textual data.

Medical AI Ethics and Safety

Her research critically addresses the ethical and safety challenges of deploying large language models in high-stakes medical domains. She focuses on creating alignment frameworks to ensure that medical LLMs are safe, trustworthy, and aligned with human values. This includes work on mitigating bias, preventing harmful outputs, and ensuring the responsible use of AI in mental healthcare, a focus that led to the development of the "Medical LLM Ethical Safety Alignment Framework" and the collaborative "Medical Health LLM Ethics and Safety White Paper."

Psychological Screening & Counseling Systems

A core part of her work involves translating research into practical applications. Her lab is dedicated to building AI-powered psychological screening and counseling systems. This involves integrating empathetic dialogue models and mental health symptom detectors into usable platforms and tools. The aim is to develop resources that can support mental health professionals and make preliminary screening and support more accessible, exemplified by her work on the "Mental Disorder Symptom Detection Model."