English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Adversarial examples are important to test and enhance the robustness of deepcode models. As source code is discrete and has to strictly stick to complexgrammar and semantics constraints, the adversarial example generationtechniques in other domains are hardly applicable. Moreover, the adversarialexample generation techniques specific to deep code models still suffer fromunsatisfactory effectiveness due to the enormous ingredient search space. Inthis work, we propose a novel adversarial example generation technique (i.e.,CODA) for testing deep code models. Its key idea is to use code differencesbetween the target input (i.e., a given code snippet as the model input) andreference inputs (i.e., the inputs that have small code differences butdifferent prediction results with the target input) to guide the generation ofadversarial examples. It considers both structure differences and identifierdifferences to preserve the original semantics. Hence, the ingredient searchspace can be largely reduced as the one constituted by the two kinds of codedifferences, and thus the testing process can be improved by designing andguiding corresponding equivalent structure transformations and identifierrenaming transformations. Our experiments on 15 deep code models demonstratethe effectiveness and efficiency of CODA, the naturalness of its generatedexamples, and its capability of enhancing model robustness after adversarialfine-tuning. For example, CODA reveals 88.05% and 72.51% more faults in modelsthan the state-of-the-art techniques (i.e., CARROT and ALERT) on average,respectively.

Code Difference Guided Adversarial Example Generation for Deep Code Models