- author = "Giovanni Pinna and Damiano Ravalico and Luigi Rovito and Luca Manzoni and Andrea {De Lorenzo}",
- title = "Improving {LLM}-Generated Code via Genetic Improvement: A Summary of Recent Advances",
- booktitle = "Ital-IA 2025: 5th National Conference on Artificial Intelligence: Thematic Workshop on Generative AI",
- year = "2025",
- editor = "Luca Manzoni and Luca Bortolussi and Giulia Cisotto and Fabio Anselmi",
- volume = "4121",
- address = "Trieste",
- month = jun # " 23-24",
- organisation = "CINI",
- publisher = "CEUR",
- keywords = "genetic algorithms, genetic programming, Genetic Improvement, Grammatical Evolution, Language Models, LLMs, Evolutionary Computation, AI, ANN, Automated program synthesis, lexicase, PSB2, HUMANEVAL, grammar-constrained GI, semantic correctness through test-driven evolution, AST, BNF, KeyBERT, Alpaca 7B/13B, LLaMA 2 13B, ChatGPT, and GPT-4, Python",
-
URL = "
https://ceur-ws.org/Vol-4121/Ital-IA_2025_paper_111.pdf",
- size = "5 pages",
- abstract = "Recent advancements in Natural Language Processing (NLP) have enabled Large Language Models (LLMs) to generate source code from textual problem descriptions. However, even state-of-the-art LLMs frequently produce syntactically correct which is, however, semantically incorrect. The LLM generated code might have a correct general structure, while treating incorrectly some corner cases. Furthermore, existing test cases might be difficult to use to help the LLM guide the construction of a semantically correct solution. To address this limitation, two recent studies propose the application of Genetic Improvement (GI) techniques, particularly Grammatical Evolution (GE), to automatically enhance the correctness of LLM-generated code. These approaches demonstrate that evolutionary computation can substantially improve the functional validity of code produced by both open-source and proprietary LLMs.",
-
notes = "Remove syntax errors in LLM output, rename variable
names
used ChatGPT
Workshop 8: Generative AI.
Department of Mathematics, Informatics and Geosciences, University of Trieste, Via Valerio 12/1, Trieste, 34127, Italy",
