Release Plan for IFEval-OOD and HREF

Thanks for the amazing work on both the OLMES and Tulu-3 releases.

As stated in Sections 7.3.1 and 7.3.2 of the Tulu-3 paper, AI2 created two benchmarks in-house, IFEval-OOD and HREF, to test the models, which still show a significant gap in optimal performance. I want to ask if the authors plan to release these two benchmarks to facilitate further research purposes. If yes, what is the potential timeline for that?

Many thanks!