A multilingual, cross-regional dataset for studying and reducing political-topic over-refusal
in open-weight language models — teaching grounded, multi-perspective, attributed
answers on politically sensitive topics. A small set of refusal rows is retained as a
diagnostic contrast, not as a coverage floor or capability objective. See
Release status and evidence limits before interpreting
this as an achieved model improvement.
License (text): CC BY… See the full description on the dataset page:
https://huggingface.co/datasets/ThePioneer/liberator-dataset.
