Our dataset consists of:

• Non-Obfuscated Code

1. C: Selected code samples from POJ-104 dataset and classic C benchmarks (Linpack, etc.);

2. JavaScript: The Octane benchmark and some web apps;

3. Python: Selected code samples from Google CodeSearchNet dataset.

• Obfuscated Code

1. Obfuscated JavaScript code (obtained by applying different obfuscation techniques to the JavaScript branch of our Non-Obfuscated Code dataset)

2. Winner code of Internet Obfuscated C Code Contest (IOCCC)

Results of our analysis:

1. Responses of different models on different code samples

2. Statistics