Traceability Link Recovery Using Retrieval Augmented Large Language Models
| Typ | Masterarbeit | |
|---|---|---|
| Aushang | TLR RA LLM.pdf | |
| Betreuer | Wenden Sie sich bei Interesse oder Fragen bitte an: Tobias Hey (E-Mail: hey@kit.edu, Telefon: +49-721-608-44765), Dominik Fuchß (E-Mail: dominik.fuchss@kit.edu, Telefon: +49-721-608-47381) |
The goal of this work is to investigate various large language models (LLMs) for their suitability in traceability link recovery (TLR) of different types of artifacts. The focus is on retrieval augmented approaches, which combine common information retrieval (IR) techniques and LLMs. The substeps of this thesis include identifying appropriate representations of the artifacts (feature engineering), developing various prompts for the LLMs (prompt engineering), and evaluating on different benchmarks.