TY - GEN
T1 - Cross-Language Code Development with Generative AI
T2 - 7th IEEE International Conference on Electronic Information and Communication Technology, ICEICT 2024
AU - Rai, Laxmisha
AU - Khatiwada, Smriti
AU - Deng, Chunrao
AU - Liu, Fasheng
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Since the release of ChatGPT in November 2022, there is growing interest around the world on exploring the capabilities of generative AI tools. In addition to text, image, audio, and video generation, these tools are also able to generate program codes. In this paper, strategies for students, programmers and enthusiasts to understand the prompting methods to generate codes in multiple languages by translating source code written in one language to another target language using generative AI is explored. The prompts are created to test the ability of generative AI to create codes in C, Java, C++, and Python. Some of the methods of generating the complete program using limited original source code statements is presented. In summary, while generating source code in a target language, generative AI tools downplay the significance of accuracy of statements written in original language, syntax, semantics, as well as missing statements in a program. Irrespective of these, generative AI tools are still able to generate complete code in a target language by correcting errors.
AB - Since the release of ChatGPT in November 2022, there is growing interest around the world on exploring the capabilities of generative AI tools. In addition to text, image, audio, and video generation, these tools are also able to generate program codes. In this paper, strategies for students, programmers and enthusiasts to understand the prompting methods to generate codes in multiple languages by translating source code written in one language to another target language using generative AI is explored. The prompts are created to test the ability of generative AI to create codes in C, Java, C++, and Python. Some of the methods of generating the complete program using limited original source code statements is presented. In summary, while generating source code in a target language, generative AI tools downplay the significance of accuracy of statements written in original language, syntax, semantics, as well as missing statements in a program. Irrespective of these, generative AI tools are still able to generate complete code in a target language by correcting errors.
KW - cross-language code
KW - Generative AI
KW - programming
KW - source code
KW - translation
UR - http://www.scopus.com/inward/record.url?scp=85205675902&partnerID=8YFLogxK
UR - https://www.mendeley.com/catalogue/2cb5c813-c60c-34f3-b4b7-63cb2a93540a/
U2 - 10.1109/ICEICT61637.2024.10671366
DO - 10.1109/ICEICT61637.2024.10671366
M3 - Conference contribution
AN - SCOPUS:85205675902
T3 - 2024 IEEE 7th International Conference on Electronic Information and Communication Technology, ICEICT 2024
SP - 562
EP - 565
BT - 2024 IEEE 7th International Conference on Electronic Information and Communication Technology, ICEICT 2024
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 31 July 2024 through 2 August 2024
ER -