DolphCoder: Echo-Locating Code Large Language Models with Diverse and Multi-Objective Instruction Tuning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913234463752192 |
|---|---|
| author | Wang, Yejie He, Keqing Dong, Guanting Wang, Pei Zeng, Weihao Diao, Muxi Mou, Yutao Zhang, Mengdi Wang, Jingang Cai, Xunliang Xu, Weiran |
| author_facet | Wang, Yejie He, Keqing Dong, Guanting Wang, Pei Zeng, Weihao Diao, Muxi Mou, Yutao Zhang, Mengdi Wang, Jingang Cai, Xunliang Xu, Weiran |
| contents | Code Large Language Models (Code LLMs) have demonstrated outstanding performance in code-related tasks. Several instruction tuning approaches have been proposed to boost the code generation performance of pre-trained Code LLMs. In this paper, we introduce a diverse instruction model (DolphCoder) with self-evaluating for code generation. It learns diverse instruction targets and combines a code evaluation objective to enhance its code generation ability. Our model achieves superior performance on the HumanEval and MBPP benchmarks, demonstrating new insights for future code instruction tuning work. Our key findings are: (1) Augmenting more diverse responses with distinct reasoning paths increases the code capability of LLMs. (2) Improving one's ability to evaluate the correctness of code solutions also enhances their ability to create it. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_09136 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | DolphCoder: Echo-Locating Code Large Language Models with Diverse and Multi-Objective Instruction Tuning Wang, Yejie He, Keqing Dong, Guanting Wang, Pei Zeng, Weihao Diao, Muxi Mou, Yutao Zhang, Mengdi Wang, Jingang Cai, Xunliang Xu, Weiran Computation and Language Artificial Intelligence Code Large Language Models (Code LLMs) have demonstrated outstanding performance in code-related tasks. Several instruction tuning approaches have been proposed to boost the code generation performance of pre-trained Code LLMs. In this paper, we introduce a diverse instruction model (DolphCoder) with self-evaluating for code generation. It learns diverse instruction targets and combines a code evaluation objective to enhance its code generation ability. Our model achieves superior performance on the HumanEval and MBPP benchmarks, demonstrating new insights for future code instruction tuning work. Our key findings are: (1) Augmenting more diverse responses with distinct reasoning paths increases the code capability of LLMs. (2) Improving one's ability to evaluate the correctness of code solutions also enhances their ability to create it. |
| title | DolphCoder: Echo-Locating Code Large Language Models with Diverse and Multi-Objective Instruction Tuning |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2402.09136 |