This is a description of the artifact of the paper “KNOD: Domain Knowledge Distilled Tree Decoder for Automated Program Repair” accepted by the ICSE 2023 technical track (paper ID: icse23main-p94), which aims to apply for available badges. We believe the artifact deserves the available badge since all the generated patches and the validated results for Defects4J v1.2, Defects4J v2.0, and QuixBugs benchmarks are publicly available. Developers can also train KNOD models and use new KNOD models to generate patches for bugs. Thus, we believe the artifact deserves the Available Badge.

In this description, we introduce how to get access to the candidate patches generated by KNOD on three benchmarks, as well as how to train new KNOD models and use KNOD models to generate patches for bugs in Defects4J v1.2, Defects4J v2.0, and QuixBugs benchmarks. We hope this artifact to benefit future work on automated program repair, and other related tasks in the software engineering community.