Many Natural Language Processing (NLP) tasks, such as sentiment analysis or syntactic parsing, have benefited from the development of word embedding models. In particular, regardless of the training algorithms, the learned embeddings have often been shown to be generalizable to different NLP tasks. In contrast, despite recent momentum on word embeddings for source code, the literature lacks evidence of their generalizability beyond the example task they have been trained for.
In this experience paper, we identify 3 potential downstream tasks, namely code comments generation, code authorship identification, and code clones detection, that source code token embedding models can be applied to. We empirically assess a recently proposed code token embedding model, namely code2vec’s token embeddings. Code2vec was trained on the task of predicting method names, and while there is potential for using the vectors it learns on other tasks, it has not been explored in literature. Therefore, we fill this gap by focusing on its generalizability for the tasks we have identified. Eventually, we show that source code token embeddings cannot be readily leveraged for the downstream tasks. Our experiments even show that our attempts to use them do not result in any improvements over less sophisticated methods. We call for more research into effective and general use of code embeddings.
Tue 12 NovDisplayed time zone: Tijuana, Baja California change
10:40 - 12:20
|Assessing the Generalizability of code2vec Token Embeddings|
Hong Jin Kang School of Information Systems, Singapore Management University, Tegawendé F. Bissyandé SnT, University of Luxembourg, David Lo Singapore Management UniversityPre-print
|Multi-Modal Attention Network Learning for Semantic Source Code Retrieval|
|Experience Paper: Search-based Testing in Automated Driving Control ApplicationsACM SIGSOFT Distinguished Paper Award|
Christoph Gladisch Corporate Research, Robert Bosch GmbH, Thomas Heinz Corporate Research, Robert Bosch GmbH, Christian Heinzemann Corporate Research, Robert Bosch GmbH, Jens Oehlerking Corporate Research, Robert Bosch GmbH, Anne von Vietinghoff Corporate Research, Robert Bosch GmbH, Tim Pfitzer Robert Bosch Automotive Steering GmbH
|Machine Translation-Based Bug Localization Technique for Bridging Lexical Gap|
Journal First Presentations
Yan Xiao Department of Computer Science, City University of Hong Kong, Jacky Keung Department of Computer Science, City University of Hong Kong, Kwabena E. Bennin Blekinge Institute of Technology, SERL Sweden, Qing Mi Department of Computer Science, City University of Hong KongLink to publication
|AutoFocus: Interpreting Attention-based Neural Networks by Code Perturbation|
Nghi D. Q. Bui Singapore Management University, Singapore, Yijun Yu The Open University, UK, Lingxiao Jiang Singapore Management UniversityPre-print
|A Quantitative Analysis Framework for Recurrent Neural Network|