Usage of Large Language Model for Code Generation Tasks: A Review

Stefano Bistarelli, Marco Fiore, Ivan Mercanti, Marina Mongiello

SN Computer Science 2025 article cited by 15

importance 4/10 partially solved Software Engineering Research software engineering stage: research mixed

The literature lacked an exhaustive analysis of the benefits and drawbacks of using large language models for generating simple and complex code.

Understanding LLM strengths and weaknesses in code generation affects software development practice, debugging, and ethical safeguards.

No AI system solved a problem; the authors performed a PRISMA-based literature review of 66 papers (2021-2023) on LLM code generation, answering 4 research questions.

Reviewed 66 papers; found Python dominance as preferred language and a significant research gap on ethical constraints.

Maps current capabilities and gaps of LLM code generation for future research.

Venue SN Computer Science
Published 2025
Work type article
Authors Stefano Bistarelli, Marco Fiore, Ivan Mercanti, Marina Mongiello
Citations 15 (OpenAlex, at import time)
References 64
Model or method GPT-4, CodeLlama (reviewed models)
Organization not stated in source
Primary source https://doi.org/10.1007/s42979-025-04241-5
Open access hybrid
Keywords large language models, code generation, PRISMA, literature review, GPT-4, CodeLlama
4 / 10

real but narrower advance

LLM code generation affects software development practice, but this is a PRISMA literature synthesis of 66 papers mapping capabilities and gaps, not a new capability itself.

Basis: inferred. Judged from the source text, which gives no explicit scale. Stated scale: not stated in source. Method and limits.

# Paper Year Citations Importance
1 ImageNet classification with deep convolutional neural networks
Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton
2017 106,345
8
2 Deep Residual Learning for Image Recognition
Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun
2016 221,134
8
3 Faster sorting algorithms discovered using deep reinforcement learning
Daniel J. Mankowitz, Andrea Michi, Anton Zhernov, Marco Gelmi, Marco S...
2023 175
8
4 A deep learning‐based approach for software vulnerability detection using code metrics
Fazli Subhan, Xiaoxue Wu, Lili Bo, Xiaobing Sun, Muhammad Rahman
2022 42
7
5 Deep Learning-Based Attack Detection and Classification in Android Devices
Alfonso Gómez, Antonio Muñoz
2023 41
7