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

Recent development of large language models (LLMs) for code like CodeX andCodeT5+ demonstrates tremendous promise in achieving code intelligence. Theirability of synthesizing code that completes a program for performing apre-defined task has been intensively tested and verified on benchmark datasetsincluding HumanEval and MBPP. Yet, evaluation of these LLMs from moreperspectives (than just program synthesis) is also anticipated, consideringtheir broad scope of applications in software engineering. In this paper, weexplore the ability of LLMs for testing programs/code. By performing thoroughanalyses of recent LLMs for code in program testing, we show a series ofintriguing properties of these models and demonstrate how program testingability of LLMs can be improved. Following recent work which utilizes generatedtest cases to enhance program synthesis, we further leverage our findings inimproving the quality of the synthesized programs and show +11.77% and +4.22%higher code pass rates on HumanEval+ comparing with the GPT-3.5-turbo baselineand the recent state-of-the-art, respectively.

The Program Testing Ability of Large Language Models for Code