MSCoT: Structured Chain-of-Thought Generation for Multiple Programming Languages

में बचाया:
ग्रंथसूची विवरण
में प्रकाशित:The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Conference Proceedings (2025), p. 1-8
मुख्य लेखक: Jin, Naizhu
अन्य लेखक: Li, Zhong, Zhang, Tian, Zeng, Qingkai
प्रकाशित:
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
विषय:
ऑनलाइन पहुंच:Citation/Abstract
टैग: टैग जोड़ें
कोई टैग नहीं, इस रिकॉर्ड को टैग करने वाले पहले व्यक्ति बनें!

MARC

LEADER 00000nab a2200000uu 4500
001 3271965835
003 UK-CbPIL
024 7 |a 10.1109/IJCNN64981.2025.11228673  |2 doi 
035 |a 3271965835 
045 2 |b d20250101  |b d20251231 
084 |a 228229  |2 nlm 
100 1 |a Jin, Naizhu  |u Nanjing University,State Key Laboratory for Novel Software Technology,China 
245 1 |a MSCoT: Structured Chain-of-Thought Generation for Multiple Programming Languages 
260 |b The Institute of Electrical and Electronics Engineers, Inc. (IEEE)  |c 2025 
513 |a Conference Proceedings 
520 3 |a Conference Title: 2025 International Joint Conference on Neural Networks (IJCNN)Conference Start Date: 2025 June 30Conference End Date: 2025 July 5Conference Location: Rome, ItalyWith the rapid development of code intelligence, the application of multiple programming languages is becoming increasingly widespread. However, most existing code generation models mainly focus on a single or a few programming languages, resulting in unsatisfactory performance in a multilingual environment. Chain-of-Thought (CoT) reasoning can significantly improve the performance of the model without the need for retraining or fine-tuning the code generation model by reasonably decomposing complex code generation tasks into multiple subtasks and gradually deriving solutions for each subtask. Nevertheless, the existing CoT generation methods mainly concentrate on Python code, and the performance on other programming languages remains unclear.To fill this gap, we first constructed a CoT generation dataset for 12 programming languages through multi-agent technology. On this basis, we proposed a CoT generation method MSCoT applicable to multiple programming languages. By introducing CoT into the code generation large model, the performance of the code generation large model in a multilingual environment can be improved. Through large-scale empirical research, we compared the generalization abilities of MSCoT and the existing CoT generation methods on multiple programming languages and proved the effectiveness of MSCoT for multiple programming languages. In addition, we also designed a human study to prove the quality of the CoT generated by MSCoT. Finally, we open-sourced the model and dataset of MSCoT to promote the research on CoT generation for multiple programming languages. 
653 |a Datasets 
653 |a Programming languages 
653 |a Performance enhancement 
653 |a Multilingualism 
653 |a Multiagent systems 
653 |a Neural networks 
653 |a Task complexity 
653 |a Economic 
700 1 |a Li, Zhong  |u Nanjing University,State Key Laboratory for Novel Software Technology,China 
700 1 |a Zhang, Tian  |u Nanjing University,State Key Laboratory for Novel Software Technology,China 
700 1 |a Zeng, Qingkai  |u Nanjing University,State Key Laboratory for Novel Software Technology,China 
773 0 |t The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Conference Proceedings  |g (2025), p. 1-8 
786 0 |d ProQuest  |t Science Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3271965835/abstract/embedded/L8HZQI7Z43R0LA5T?source=fedsrch