The IRSE track consists of two tasks:
Task1: Generative AI-based Software Metadata Classification
A binary code comment quality classification model needs to be augmented with generated code and comment pairs that can improve the model's accuracy.
Task 2: Translation of Programming Language to Natural Language
Code summaries (intents) are to be generated by training over 40,000 code and intent pairs written in python.
Important Dates
10th June 2023 - Training data release
3rd August , 2023 – Test data release
30th August, 2023 – Run submission deadline
3rd September , 2023 – Results declared
15th September, 2023 – Working notes due (4 page 2 column)
15th October, 2023 – Camera ready copies of working notes