Matches in SemOpenAlex for { <https://semopenalex.org/work/W4367000497> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W4367000497 abstract "Transformer-based pre-trained models have recently achieved great results in solving many software engineering tasks including automatic code completion which is a staple in a developer's toolkit. While many have striven to improve the code-understanding abilities of such models, the opposite -- making the code easier to understand -- has not been properly investigated. In this study, we aim to answer whether making code easier to understand through using contextual data improves the performance of pre-trained code language models for the task of code completion. We consider type annotations and comments as two common forms of additional contextual information that often help developers understand code better. For the experiments, we study code completion in two granularity levels; token and line completion and take three recent and large-scale language models for source code: UniXcoder, CodeGPT, and InCoder with five evaluation metrics. Finally, we perform the Wilcoxon Signed Rank test to gauge significance and measure the effect size. Contrary to our expectations, all models perform better if type annotations are removed (albeit the effect sizes are small). For comments, we find that the models perform better in the presence of multi-line comments (again with small effect sizes). Based on our observations, we recommend making proper design choices when training, fine-tuning, or simply selecting such models given the intended data and application. Better evaluations and multi-modal techniques can also be further investigated to improve the practicality and accuracy of auto-completions." @default.
- W4367000497 created "2023-04-27" @default.
- W4367000497 creator A5064355563 @default.
- W4367000497 creator A5074259780 @default.
- W4367000497 creator A5090401584 @default.
- W4367000497 date "2023-04-24" @default.
- W4367000497 modified "2023-09-27" @default.
- W4367000497 title "Enriching Source Code with Contextual Data for Code Completion Models: An Empirical Study" @default.
- W4367000497 doi "https://doi.org/10.48550/arxiv.2304.12269" @default.
- W4367000497 hasPublicationYear "2023" @default.
- W4367000497 type Work @default.
- W4367000497 citedByCount "0" @default.
- W4367000497 crossrefType "posted-content" @default.
- W4367000497 hasAuthorship W4367000497A5064355563 @default.
- W4367000497 hasAuthorship W4367000497A5074259780 @default.
- W4367000497 hasAuthorship W4367000497A5090401584 @default.
- W4367000497 hasBestOaLocation W43670004971 @default.
- W4367000497 hasConcept C119857082 @default.
- W4367000497 hasConcept C124101348 @default.
- W4367000497 hasConcept C137287247 @default.
- W4367000497 hasConcept C137293760 @default.
- W4367000497 hasConcept C150292731 @default.
- W4367000497 hasConcept C154945302 @default.
- W4367000497 hasConcept C177264268 @default.
- W4367000497 hasConcept C177774035 @default.
- W4367000497 hasConcept C199360897 @default.
- W4367000497 hasConcept C199519371 @default.
- W4367000497 hasConcept C2776760102 @default.
- W4367000497 hasConcept C2777904410 @default.
- W4367000497 hasConcept C41008148 @default.
- W4367000497 hasConcept C43126263 @default.
- W4367000497 hasConcept C529173508 @default.
- W4367000497 hasConceptScore W4367000497C119857082 @default.
- W4367000497 hasConceptScore W4367000497C124101348 @default.
- W4367000497 hasConceptScore W4367000497C137287247 @default.
- W4367000497 hasConceptScore W4367000497C137293760 @default.
- W4367000497 hasConceptScore W4367000497C150292731 @default.
- W4367000497 hasConceptScore W4367000497C154945302 @default.
- W4367000497 hasConceptScore W4367000497C177264268 @default.
- W4367000497 hasConceptScore W4367000497C177774035 @default.
- W4367000497 hasConceptScore W4367000497C199360897 @default.
- W4367000497 hasConceptScore W4367000497C199519371 @default.
- W4367000497 hasConceptScore W4367000497C2776760102 @default.
- W4367000497 hasConceptScore W4367000497C2777904410 @default.
- W4367000497 hasConceptScore W4367000497C41008148 @default.
- W4367000497 hasConceptScore W4367000497C43126263 @default.
- W4367000497 hasConceptScore W4367000497C529173508 @default.
- W4367000497 hasLocation W43670004971 @default.
- W4367000497 hasOpenAccess W4367000497 @default.
- W4367000497 hasPrimaryLocation W43670004971 @default.
- W4367000497 hasRelatedWork W1534236646 @default.
- W4367000497 hasRelatedWork W167779242 @default.
- W4367000497 hasRelatedWork W2134091519 @default.
- W4367000497 hasRelatedWork W2149635184 @default.
- W4367000497 hasRelatedWork W2527612534 @default.
- W4367000497 hasRelatedWork W3014307249 @default.
- W4367000497 hasRelatedWork W3025557260 @default.
- W4367000497 hasRelatedWork W3080544106 @default.
- W4367000497 hasRelatedWork W3197188256 @default.
- W4367000497 hasRelatedWork W4250385364 @default.
- W4367000497 isParatext "false" @default.
- W4367000497 isRetracted "false" @default.
- W4367000497 workType "article" @default.