
Registered user since Fri 25 Jan 2019
Tobias Hey is a postdoctoral researcher in the Modelling for Continuous Software Engineering group at the Karlsruhe Institute of Technology (KIT), focusing on the intersection of software engineering and natural language processing (NLP). His research centers on traceability link recovery, requirements analysis, and the integration of machine learning and large language models (LLMs) into software engineering processes.
He has proposed techniques such as FTLR (Fine-grained Traceability Link Recovery) and NoRBERT, the first LLM-based classifier for requirements, to improve the automation and accuracy of traceability in software systems. His recent work includes LiSSA, a retrieval-augmented generation (RAG) approach for generic traceability link recovery, and prior research on leveraging intermediate models to recover trace links between documentation and code.
Tobias regularly publishes at venues such as ICSE, RE, REFSQ, and ICSA, with a focus on scalable, explainable, and high-precision approaches to software traceability and requirements engineering.
Contributions