Data: Large Language Models Require Curated Context for Reliable Political Fact-Checking—Even with Reasoning and Web Search

Data for the following research paper:

Title: Large Language Modles Require Curated Context for Reliable Political Fact-Checking—Even with Reasoning and Web Search

Code: https://github.com/osome-iu/fact_check_rag_osome

Authors: Matthew R. DeVerna, Kai-Cheng Yang, Harry Yaojun Yan, Filippo Menczer

---

These files are intended to be downloaded and used with the code repository linked above. They’re too large to include directly on GitHub.

---

Unpacking the data tarball
--------------------------

This repository includes a compressed archive named data.tar.gz that contains all required data files.

First, clone the GitHub repository linked above. Second, download the data.tar.gz file stored here. Then, move it into the project root directory of the cloned GitHub repository.

Then extract the data from the archive, by running the following command from the project root:

• tar -xzvf data.tar.gz

This will recreate the data/ directory and its full contents.

Additional notes:

• If you are on Windows without WSL, you may need a tool such as Git Bash, 7-Zip, WinRAR, or Windows Subsystem for Linux (WSL) to unpack the tarball.
• Once extracted, see the README files for details about all files
• Once extracted, all data represents approximately 5.3GBs