[NeurIPS 2025] The official implementation of "RF-Agent: Automated Reward Function Design via Language Agent Tree Search"

We are currently reorganising the project page.

In the meantime, supplementary materials in OpenReview contain offline code which you may refer to for now.