DolphCoder: Echo-Locating Code Large Language Models with Diverse and Multi-Objective Instruction Tuning
Authors: Wang, Yejie and He, Keqing and Dong, Guanting and Wang, Pei and Zeng, Weihao and Diao, Muxi and Xu, Weiran and Wang, Jingang and Zhang, Mengdi and Cai, Xunliang
Abstract:
Code Large Language Models (Code LLMs) have demonstrated outstanding performance in code-related tasks. Various instruction finetuning 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 more distinct reasoning paths increases the code capability of LLMs. (2) Improving one’s ability to evaluate the correctness of code also enhances their ability to create it.
Link: Read Paper
Labels: code generation, program synthesis, code model, code model training, source code model