Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

line-level code completion

Line-level code completion is an automated software development task in which a machine learning model or code intelligence system predicts and generates the remainder or entirety of a single line of source code based on preceding context. Positioned between fine-grained single-token suggestions and broader multi-line or function-level code synthesis, this technique focuses on completing discrete, syntactically coherent statements within an active file. Typically powered by generative neural networks and pre-trained language models, line-level completion tools analyze contextual information such as prior lines, programmatic scope, and repository structure to suggest appropriate syntax, identifiers, function calls, and arguments, thereby accelerating the programming process and minimizing routine syntax errors.

1 item

CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generation

CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generation

Shuai Lu, Daya Guo, Shuo Ren, Junjie Huang, Alexey Svyatkovskiy, Ambrosio Blanco, Colin Clement, Dawn Drain, Daxin Jiang, Duyu Tang, Ge Li, Lidong Zhou, Linjun Shou, Long Zhou, Michele Tufano, Ming Gong, Ming Zhou, Nan Duan, Neel Sundaresan, Shao Kun Deng, Shengyu Fu, Shujie Liu

OrganizationsBeihang UniversityMicrosoftPeking UniversitySun Yat-sen University

Why you should read this

Presents a standardized benchmark and evaluation platform spanning 10 code understanding and generation tasks across 14 datasets to systematically measure and advance machine learning models for software engineering.

Benchmark datasets have a significant impact on accelerating research in programming language tasks. In this paper, we introduce CodeXGLUE, a benchmark dataset to foster machine learning research for program understanding and generation. CodeXGLUE includes a collection of 10 tasks across 14 datasets and a platform for model evaluation and comparison. CodeXGLUE also features three baseline systems, including the BERT-style, GPT-style, and Encoder-Decoder models, to make it easy for researchers to use the platform. The availability of such data and baselines can help the development and validation of new methods that can be applied to various program understanding and generation problems.

Added

2026-09-24