Supplementary Material of ST RK: Benchmarking LLM Retrieval on Textual and Relational Knowledge Bases

Code: We release a PyPI package, stark-qa (https://pypi.org/project/stark-qa/). The Croissant metadata for our dataset is available for viewing and downloading at https://stark.stanford.edu/files/croissant_metadata.json. We provide a persistent dereferenceable identifier DOI: https://doi.org/10.57967/hf/2530. RK retrieval datasets are under license CC-BY-4.0 as stated in our website. And our released code is under MIT license, as stated in the GitHub repository. We plan to update our website with the most recent document and Python package. We will maintain our GitHub repository will pull requests and open issues. We hereby confirm that we bear all responsibility for any violation of rights that may occur in the use or distribution of the data and content presented in this work. We affirm that we have obtained all necessary permissions and licenses for the data and content included in this work. We confirm that the use of this data complies with all relevant laws and regulations, and we take full responsibility for addressing any claims or disputes that may arise regarding rights violations or licensing issues.