How Do Your Code LLMs Perform? Empowering Code Instruction Tuning with High-Quality Data

本研究探讨了如何通过高质量数据来提升代码语言模型的性能。我们提出了一种新的方法来增强代码指令调优，并开源了相关代码：https://github.com/banksy23/XCoder

Recommended citation: Wang, Y.*, He, K.*, Fu, D.*, Gongque, Z., Xu, H., Chen, Y., Wang, Z., Fu, Y., Dong, G., Diao, M., Wang, J., Zhang, M., Cai, X., & Xu, W. (2024). How Do Your Code LLMs Perform? Empowering Code Instruction Tuning with High-Quality Data. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing.
Download Paper