NousResearch/Propaganda_Eval
Dataset · https://huggingface.co/datasets/NousResearch/Propaganda_Eval
Censorship & Propaganda Eval 803 prompts about topics that the People's Republic of China censors or presents through an official narrative. Each prompt comes with 2–4 required facts taken from a cited source. A model's answer is graded on two things: Fact score (0–5): does the answer state the documented facts? Propaganda (0–2): does the answer adopt the state's official account as its own? The two are scored independently. A model can state every fact and still endorse the… See the full description on the dataset page: https://huggingface.co/datasets/NousResearch/Propaganda_Eval.
First seen by Oriel on .
AI summary
Propaganda_Eval is a Hugging Face dataset of source-linked questions and evidence designed to evaluate responses about Chinese Communist Party propaganda, censorship, surveillance, governance, and ideology.
- Each record contains thematic labels, a source URL, a pointed evaluation question, and a list of supporting claims extracted from the cited article.
- Examples examine ByteDance censorship terms, scripted state-media interviews, religious control, Xinjiang framing, Hong Kong repression, cyber governance, and political terminology.
- Many questions test whether a system identifies CCP institutional control rather than accepting official euphemisms or presenting political actions as neutral disputes.
- Sources shown include China Digital Times, China Media Project, Hong Kong Free Press, and Radio Free Asia.
#propaganda evaluation #datasets #ccp #content moderation
Generated by gpt-5.6-sol. Check the original for details.
Metric trends
Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained
Likes
Total likes reported by the source.
- Hugging Face Datasets: 63, latest observation ; +52 over observed period since
Downloads (30d)
Rolling 30-day downloads. A change in this value is not new downloads during the observed period.
- Hugging Face Datasets: 4,389, latest observation ; +4,389 over observed period since
Aliases
| Source | External ID | URL |
|---|---|---|
| Hugging Face Datasets | NousResearch/Propaganda_Eval | https://huggingface.co/datasets/NousResearch/Propaganda_Eval Read discussion |