Upload README.md with huggingface_hub

README.md

@@ -40,10 +40,9 @@ All documents were annotated by Llama 4 Maverick (meta-llama/Llama-4-Maverick-17
 
 ## Key findings
 
-- **4.1% of random FineWeb-Edu documents** contain misinformation (health misinformation, pseudoscience, conspiracy theories, etc.)
-- **39% of documents from known problematic domains** contain misinformation content
-- Problematic content scores well on the edu-score classifier because it mimics academic writing style
-- The full FineWeb-Edu dataset contains 1.53 billion documents. Extrapolating the 4.1% background rate, approximately **63 million documents** may contain misinformation. From known problematic domains alone, an estimated **3 million documents** contain confirmed misinformation content.
+- **4.1% of randomly sampled FineWeb-Edu documents** contain misinformation (health misinformation, pseudoscience, conspiracy theories, etc.)
+- **39% of documents from known problematic domains** contain misinformation content, nearly 10x the overall rate
+- The full FineWeb-Edu dataset contains 1.53 billion documents. Extrapolating the 4.1% rate, approximately **63 million documents** may contain misinformation. From known problematic domains alone, an estimated **3 million documents** contain confirmed misinformation content.
 
 ## Label distribution