End-to-end Log Statement Generation at Block-Level
Journal of Systems and Software (JSS)(2024)CCF BSCI 2区
Chongqing Univ
Abstract
Logging is crucial in software development for addressing runtime issues but can pose challenges. Logging encompasses four essential sub-tasks: whether to log (Whether), where to log (Position), which log level (Level), and what information to log (Message). While existing approaches have performed well, they suffer from two limitations. Firstly, they address only a subset of the logging sub-tasks. Secondly, most of them focus on generating single log statements at class or method level, potentially overlooking multiple log statements within those scopes.To address these issues, we propose ELogger, which enables end-to-end log statement generation at block-level. Furthermore, ELogger implements block-level log generation, enabling it to handle multiple log statements within different code blocks of a method. Evaluation results indicate that ELogger correctly predicts all four sub-tasks in 19.55% of cases. Compared to the baselines that combined existing approaches for end-to-end log statement generation, ELogger demonstrates a significant improvement with a 50.85% to 78.21% average increase. Additionally, ELogger correctly predicts whether to log in 71.68% of cases, two sub-tasks (Whether and Position) in 58.29% of cases, and three sub-tasks (Whether, Position, and Level) in 41.97% of cases, all of which outperform the baselines.
MoreTranslated text
Key words
Log statement,End-to-end,Block-level,Deep learning
求助PDF
上传PDF
View via Publisher
AI Read Science
AI Summary
AI Summary is the key point extracted automatically understanding the full text of the paper, including the background, methods, results, conclusions, icons and other key content, so that you can get the outline of the paper at a glance.
Example
Background
Key content
Introduction
Methods
Results
Related work
Fund
Key content
- Pretraining has recently greatly promoted the development of natural language processing (NLP)
- We show that M6 outperforms the baselines in multimodal downstream tasks, and the large M6 with 10 parameters can reach a better performance
- We propose a method called M6 that is able to process information of multiple modalities and perform both single-modal and cross-modal understanding and generation
- The model is scaled to large model with 10 billion parameters with sophisticated deployment, and the 10 -parameter M6-large is the largest pretrained model in Chinese
- Experimental results show that our proposed M6 outperforms the baseline in a number of downstream tasks concerning both single modality and multiple modalities We will continue the pretraining of extremely large models by increasing data to explore the limit of its performance
Upload PDF to Generate Summary
Must-Reading Tree
Example

Generate MRT to find the research sequence of this paper
Related Papers
2013
被引用79 | 浏览
2016
被引用88 | 浏览
2017
被引用65 | 浏览
2019
被引用344 | 浏览
2021
被引用67 | 浏览
2023
被引用77 | 浏览
2023
被引用7 | 浏览
2023
被引用12 | 浏览
Data Disclaimer
The page data are from open Internet sources, cooperative publishers and automatic analysis results through AI technology. We do not make any commitments and guarantees for the validity, accuracy, correctness, reliability, completeness and timeliness of the page data. If you have any questions, please contact us by email: report@aminer.cn
Chat Paper
Summary is being generated by the instructions you defined