Aims and scope of HEALxLM

Recent advancements in generative AI have paved the way for powerful Language Model techniques to be applied in health-related contexts, offering potential benefits for healthcare providers and patients. These models can support medical professionals by aiding diagnosis, treatment planning, and process optimisation, offering patients more accessible and immediate guidance. However, several challenges must be addressed to ensure safe and effective use. Key issues include the accuracy and reliability of medical information, as any misinformation could have serious consequences. Additionally, protecting patient privacy and data security is a priority, as sensitive health data must be handled according to the reference regulations. Addressing biases in AI-generated responses is another critical concern, as ensuring that sensitive topics, such as mental health or end-of-life care, are approached with empathy and professionalism. Moreover, for AI models to be truly effective in clinical settings, They must be integrated with existing healthcare systems and continuously updated to reflect the latest medical knowledge and guidelines.While much discussion surrounds the application of Language models in healthcare, their integration into clinical practice is complex and unresolved. The Healthcare Empowerment and Advancement Leveraging xLM (HEALxLM)  will, therefore, explore these challenges and potential solutions for advancing language models, both small and large (xLM), for applications in healthcare.

Topics

More specifically, topics of interest include, but are not limited to:

• Language Models for Home and Remote care   
  • Chatbot-based and virtual assistant solution for supporting patient self-management
• Multimodality in Healthcare   
  • Integrating text, speech, and image data for comprehensive multi-perspective xLM diagnostic systems
  • Multimodal data to improve healthcare services based on xLMs
• Clinical Decision Support   
  • Novel xLM-based solutions to assist healthcare professionals
• Trustworthy AI in Healthcare   
  • RAG-based solutions and fine-tuning for improved xLM performances
• Real-world Applications   
  • Real-world applications that exploit xLMs   
  • xLMs Systems for Big Data scenarios   
  • xLMs for a diverse, equitable, and inclusive society

Submission instructions

Authors are invited to submit full papers that are unpublished and not under review elsewhere. Authors are invited to submit full papers that are unpublished and not under review elsewhere. All submissions will undergo a peer-review process by Program Committee members in a single blind fashion. Please look at the official site to more details: https://healxlm.github.io/2025/cfp/

Organization Chairs

• Sara Montagna, Department of Pure and Applied Science (DiSPeA) - University of Urbino Carlo Bo, Italy

• Gianluca Aguzzi, Department of Computer Science and Engineering (DISI) - University of Bologna, Italy

• Mauro Dragoni, Fondazione Bruno Kessler (FBK), Italy