Method name suggestion with hierarchical attention networks

Xu, Sihan; Zhang, Sen; Wang, Weijing; Cao, Xinya; Guo, Chenkai; Xu, Jing

doi:10.1145/3294032.3294079

Cited by 27 publications

(18 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Allamanis et al [3] used logbilinear neural language models supplemented by additional manual features to predict Java method and class names. Java method names have also been treated as short, descriptive "summaries" of its body; in this view, prior work has augmented attention mechanisms in convolutional networks [5], used sequence-to-sequence models to learn from descriptions (e.g., Javadoc comments) [27], and utilized the tree-structure of the code in a hierarchical attention network [74]. Unlike Java syntax trees, Coq syntax and kernel trees contain considerable semantic information useful for naming.…”

Section: Related Workmentioning

confidence: 99%

“…However, serious challenges are posed by, on the one hand, Coq's powerful language extension facilities and fusion of type checking and computation [12], and on the other hand, the idiosyncratic conventions used by Coq practitioners compared to software engineers. Hence, although suggesting lemma names is similar in spirit to suggesting method names in Java-like languages [74], the former task is more challenging in that lemma names are typically much shorter than method names and tend to include heavily abbreviated terminology from logic and advanced mathematics; a single character can carry significant information about a lemma's meaning. For example, the MathComp lemma names card support normedTI ("cardinality of support groups of a normed trivial intersection group") and extprod mulgA ("associativity of multiplication operations in external product groups") concisely convey information on lemma statement structure and meaning through both abbreviations and suffixes, as when the suffix A indicates an associative property.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Generation of Coq Lemma Names Using Elaborated Terms

Nie

Palmskog

et al. 2020

Automated Reasoning

View full text Add to dashboard Cite

Coding conventions for naming, spacing, and other essentially stylistic properties are necessary for developers to effectively understand, review, and modify source code in large software projects. Consistent conventions in verification projects based on proof assistants, such as Coq, increase in importance as projects grow in size and scope. While conventions can be documented and enforced manually at high cost, emerging approaches automatically learn and suggest idiomatic names in Java-like languages by applying statistical language models on large code corpora. However, due to its powerful language extension facilities and fusion of type checking and computation, Coq is a challenging target for automated learning techniques. We present novel generation models for learning and suggesting lemma names for Coq projects. Our models, based on multi-input neural networks, are the first to leverage syntactic and semantic information from Coq's lexer (tokens in lemma statements), parser (syntax trees), and kernel (elaborated terms) for naming; the key insight is that learning from elaborated terms can substantially boost model performance. We implemented our models in a toolchain, dubbed Roosterize, and applied it on a large corpus of code derived from the Mathematical Components family of projects, known for its stringent coding conventions. Our results show that Roosterize substantially outperforms baselines for suggesting lemma names, highlighting the importance of using multi-input models and elaborated terms.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Deep Generation of Coq Lemma Names Using Elaborated Terms

Nie

Palmskog

et al. 2020

Automated Reasoning

View full text Add to dashboard Cite

show abstract

“…For the code suggestion area, other research works focuses on a specific types of code tokens. [49] suggested method name based on Hierachical Attention Networks, and [5] suggested method name and class name. In our work, we intend to generate all types of tokens based on writing the abbreviations or prefixes.…”

Section: Related Workmentioning

confidence: 99%

Statistical machine translation outperforms neural machine translation in software engineering: why and how

Phan

Jannesari

2020

Proceedings of the 1st ACM SIGSOFT International Workshop on Representation Learning for Software Engineering and Program Langu

View full text Add to dashboard Cite

Neural Machine Translation (NMT) is the current trend approach in Natural Language Processing (NLP) to solve the problem of automatically inferring the content of target language given the source language. The ability of NMT is to learn deep knowledge inside languages by deep learning approaches. However, prior works show that NMT has its own drawbacks in NLP and in some research problems of Software Engineering (SE). In this work, we provide a hypothesis that SE corpus has inherent characteristics that NMT will confront challenges compared to the state-of-the-art translation engine based on Statistical Machine Translation. We introduce a problem which is significant in SE and has characteristics that challenges the ability of NMT to learn correct sequences, called Prefix Mapping. We implement and optimize the original SMT and NMT to mitigate those challenges. By the evaluation, we show that SMT outperforms NMT for this research problem, which provides potential directions to optimize the current NMT engines for specific classes of parallel corpus. By achieving the accuracy from 65% to 90% for code tokens generation of 1000 Github code corpus, we show the potential of using MT for code completion at token level.

show abstract

“…In this paper, we proposed several extensions to code2seq. Xu et al (2019) used a hierarchical attention network for function name generation. In this model, the important information of the lower layer is passed to the upper layer by a recursive network.…”

Section: Outputs Of Each Modelmentioning

confidence: 99%

Pointing to Subwords for Generating Function Names in Source Code

Fujita

Kamigaito

Takamura

et al. 2020

Proceedings of the 28th International Conference on Computational Linguistics

View full text Add to dashboard Cite

We tackle the task of automatically generating a function name from source code. Existing generators face difficulties in generating low-frequency or out-of-vocabulary subwords. In this paper, we propose two strategies for copying low-frequency or out-of-vocabulary subwords in inputs. Our best performing model showed an improvement over the conventional method in terms of our modified F1 and accuracy on the Java-small and Java-large datasets.

show abstract

Method name suggestion with hierarchical attention networks

Cited by 27 publications

References 37 publications

Deep Generation of Coq Lemma Names Using Elaborated Terms

Deep Generation of Coq Lemma Names Using Elaborated Terms

Statistical machine translation outperforms neural machine translation in software engineering: why and how

Pointing to Subwords for Generating Function Names in Source Code

Contact Info

Product

Resources

About