Research & Projects
Research Topics
- Responsible AI, fairness, social bias, explainability, transparency
- Large Language Models, text embeddings, knowledge representations, semantics
- Software Engineering for AI, responsible design, requirements, testing
- Big Data Analytics, data management, edge computing, federated learning
Research Vision
My research vision is about achieving reliable and trustworthy AI for the social good, with a focus on LLMs and AI-based systems. My interests include social biases in language models from text embeddings to LLMs, fairness evaluation, and mitigation strategies.
Currently, I am researching strategies for reducing LLM hallucinations by knowledge graph integration as a mechanism for grounding model outputs and improving factual accuracy and consistency.
I also worked on responsible software engineering for AI-based software systems, at the intersection of requirements engineering, software testing, and AI ethics. I advocate for a multidisciplinary approach to responsible AI, in line with the Digital Humanism initiative.
Finally, I have a background in big data architectures and data management, including edge computing, federated learning, and federated architectures for data analytics.
