Hello,
Thanks for releasing the GOAT-Bench code and dataset. The modular approaches are a core part of the benchmark’s comparisons and analyses, so having their code would be essential for reproducing the results and building upon the work. Could you clarify:
- Are the modular model implementations available somewhere else?
- If not, are there plans to release them (and if so, an approximate timeline)?
Thanks for your efforts in creating and sharing this benchmark — it’s a valuable resource for the community.
Hello,
Thanks for releasing the GOAT-Bench code and dataset. The modular approaches are a core part of the benchmark’s comparisons and analyses, so having their code would be essential for reproducing the results and building upon the work. Could you clarify:
Thanks for your efforts in creating and sharing this benchmark — it’s a valuable resource for the community.