R&D-Agent paper question

[alexholdenmiller]

❓ Questions and Help

Hi there!

I really enjoyed reading your paper "R&D-Agent: Automating Data-Driven AI Solution Building Through LLM-Powered Automated Research, Development, and Evolution" (https://arxiv.org/html/2505.14738v1). It's a great step forward in agents that can automatically solve data science tasks.

I am interested in which prompts you used for the paper.

For example, are these the ones you used for the kaggle tasks in MLE-Bench?
https://github.com/microsoft/RD-Agent/blob/main/rdagent/scenarios/kaggle/prompts.yaml

Or maybe this one?
https://github.com/microsoft/RD-Agent/blob/main/rdagent/scenarios/data_science/proposal/exp_gen/prompts_v3.yaml

Would love to better understand your work, thank you in advance!

[you-n-g]

Thanks for your attention.

Most of the prompts are under this sub directry.
https://github.com/microsoft/RD-Agent/blob/main/rdagent/scenarios/data_science/

Most of current experiments are based on prompts in this file.
https://github.com/microsoft/RD-Agent/blob/main/rdagent/scenarios/data_science/proposal/exp_gen/prompts_v2.yaml