Publications

You can find all my papers on my Google Scholar

(*: corresponding author)

LLM Reasoning and Adaptation

General methods for improving how large language models reason and adapt, so they learn efficiently, reason reliably, and generalize across tasks and domains.

• Mental-R1: Aligning LLM Reasoning for Mental Health Assessment.
  Xin Wang, Boyan Gao, Yibo Yang, David A Clifton
  Preprint, arXiv:2606.13176 [pdf]

• Optimization-Inspired Few-Shot Adaptation for Large Language Models.
  Boyan Gao, Xin Wang, Yibo Yang, David A Clifton
  Advances in Neural Information Processing Systems (NeurIPS 2025, Spotlight) [pdf]

AI for Mental Health

Building AI systems that understand mental health at progressively finer levels of granularity, from representation learning to stress recognition, stress category classification, and stressor identification, across language, image, and behavioral signals.

• MISE: Meta-knowledge Inheritance for Social Media-Based Stressor Estimation.
  Xin Wang, Ling Feng, Huijun Zhang, Lei Cao, Kaisheng Zeng, Qi Li, Yang Ding, Yi Dai, David A Clifton
  Proceedings of the ACM Web Conference (WWW 2025, Oral) [pdf][code][data]

• Contrastive Learning of Stress-specific Word Embedding for Social Media-based Stress Detection.
  Xin Wang, Huijun Zhang, Lei Cao, Kaisheng Zeng, Qi Li, Ningyun Li, Ling Feng
  Proceedings of the ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023) [pdf][code][data]

• A Meta-learning based Stress Category Detection Framework on Social Media.
  Xin Wang, Lei Cao, Huijun Zhang, Ling Feng, Yang Ding, Ningyun Li
  Proceedings of the ACM Web Conference (WWW 2022) [pdf][data]

• Integrating Content-Semantics-World Knowledge to Detect Stress from Videos.
  Yang Ding, Yi Dai, Xin Wang*, Ling Feng*, Lei Cao, Huijun Zhang
  Proceedings of the ACM International Conference on Multimedia (MM 2024) [pdf]

• Leverage Social Media for Personalized Stress Detection.
  Xin Wang, Huijun Zhang, Lei Cao, Ling Feng
  Proceedings of the ACM International Conference on Multimedia (MM 2020) [pdf]

AI for Biomedicine

AI methods for biomedical discovery, spanning representation learning, knowledge graph reasoning, and predictive modeling over complex biological and medical data, including drug–target interaction and cold-start settings.

• IPM-DTI: An Interaction-Pattern-Driven Multimodal Framework for Drug–Target Interaction Prediction via Knowledge Graphs.
  Yao Liu, Yifei Zhou, Xin Wang, Lei Zhao
  Journal of Chemical Information and Modeling (JCIM 2026) [pdf]

• MuRL-DTI: A Multimodal Feature Fusion Reinforcement Learning Approach for Cold Start in Drug–Target Interactions.
  Yao Liu, Xin Wang, Ye Liu, Dandan Dou
  IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025) [pdf]

• SAGS-DynamicBio: Integrating Semantic-Aware and Graph Structure-Aware Embedding for Dynamic Biological Data with Knowledge Graphs.
  Yao Liu, Yongfei Zhang, Xin Wang
  European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD 2024) [pdf]