As a doctoral researcher at the Software Technology Group, my goal is to improve code intelligence by utilizing code graphs and multimodality techniques. I have worked on natural language processing projects. I have also worked on image generation and editing with textual input to make such multimodal models better steerable and explainable (see my scholar page).
Currently, I am investigating generative models for code. This includes creating robust and secure code models that has better code understanding them using multimodal techniques where I can utilize my previous experiences with NLP and vision-language models.
If you are interested in theses or projects that are related to code intelligence and multimodality, feel free to contact me.