Challenges to Using Large Language Models in Code Generation and Repair
Large language models (LLMs) hold great promise in solving many challenges arising from software complexity, including the possibility of automating code generation and repair. Although we cannot deny the groundbreaking nature of LLM-based code repair, we must be realistic in positioning current results.
University College Dublin, Dublin, Ireland
SAP, Mougins, France
North Carolina State University, Raleigh, NC, USA
Department of Software Engineering, University of Szeged, Szeged, Hungary
University of Hawaii at Manoa, Honolulu, HI, USA
Massachusetts Institute of Technology (MIT) Lincoln Laboratory, Lexington, MA, USA
École Polytechnique Fédérale de Lausanne
University of Bristol, Bristol, U.K.
Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, USA
Cybersecurity and Infrastructure Security Agency, VA, USA
2025-03
23
2
81
88
REVIEWED
EPFL