Papers › Better Language Models of Code through Self-Improvement

Better Language Models of Code through Self-Improvement

2 Apr 2023arXiv:2304.01228archive 2025-07-28

Hung Quoc To, Nghi D. Q. Bui, Jin Guo, Tien N. Nguyen

Pre-trained language models for code (PLMCs) have gained attention in recent research. These models are pre-trained on large-scale datasets using multi-modal objectives. However, fine-tuning them requires extensive supervision and is limited by the size of the dataset provided. We aim to improve this issue by proposing a simple data augmentation framework. Our framework utilizes knowledge gained during the pre-training and fine-tuning stage to generate pseudo data, which is then used as training data for the next step. We incorporate this framework into the state-of-the-art language models, such as CodeT5, CodeBERT, and UnixCoder. The results show that our framework significantly improves PLMCs' performance in code-related sequence generation tasks, such as code summarization and code generation in the CodeXGLUE benchmark.

PaperPDFCode

Code

Fsoft-AIC/Code-LM-Self-Improvement mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Code SummarizationData Augmentation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AdafactorAttentionAttention DropoutBPECodeBERTDense ConnectionsDropoutGated Linear UnitInverse Square Root ScheduleLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSentencePieceSoftmaxT5

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections