Data generation

[keven-cyber]

Thank you very much for your excellent work and the improvements compared to the previous benchmark. May I ask if the code allows me to easily generate data myself?

[hewei2001]

Thank you so much for your interest in our work! 🙏

We're glad to hear that you appreciate the improvements in our benchmark. Regarding your question about generating custom data: currently, the data in our benchmark is generated using a model-assisted pipeline with human refinement. Due to company policies, the code for data construction isn’t available for open-sourcing at the moment.

If you'd like to create similar data, we recommend following the pipeline described in our paper. We appreciate your understanding and would be happy to answer any further questions you might have!