Answers are grounded in this paper and cite their sources.
We report on the organization and results of the tool competition of the fifth International Workshop on Natural Language-based Software Engineering (NLBSE'26). As in prior editions, we organized the competition on automated code comment classification, with a larger dataset and we introduced the skill classification competition. In this tool competition edition, five teams submitted classification models to automatically classify code comments. The submitted models were fine-tuned and evaluated on a benchmark dataset of 9,361 code comments. This paper reports the competition rules, the accepted teams and submitted models, and the evaluation framework for the code-comment classification and skill-classification competitions, with submissions ranked based on the models' predictive performance and inference efficiency.
No takes yet. Share an insight, caveat, or question.
Mock et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: