1. Join the VAKRA Advanced mailing list from the original task site: https://sites.google.com/view/vakrasemeval2027
2. Read the public benchmark setup guide at https://github.com/IBM/vakra/blob/main/setup.md. It covers cloning the repository, preparing a Python environment, downloading training data, and running the local environment.
3. Explore the training dataset at https://huggingface.co/datasets/ibm-research/VAKRA and use the repository to develop and evaluate your agent.
4. Register to access the evaluation dataset using the following form https://forms.gle/2MLpxaU1kcD1SbCm7