Update README.md

README.md

@@ -33,7 +33,7 @@ We also have other three MiniCheck model variants:
 </p>
 
 The performance of these models is evaluated on our new collected benchmark (unseen by our models during training), [LLM-AggreFact](https://huggingface.co/datasets/lytang/LLM-AggreFact), 
-from 10 recent human annotated datasets on fact-checking and grounding LLM generations. MiniCheck-RoBERTa-Large outperform all
+from 11 recent human annotated datasets on fact-checking and grounding LLM generations. MiniCheck-RoBERTa-Large outperform all
 exisiting specialized fact-checkers with a similar scale by a large margin. See full results in our work.
 
 Note: We only evaluated the performance of our models on real claims -- without any human intervention in 
@@ -73,7 +73,7 @@ from minicheck.minicheck import MiniCheck
 import os
 os.environ["CUDA_VISIBLE_DEVICES"] = "0"
 
-# load 13K test data
+# load 29K test data
 df = pd.DataFrame(load_dataset("lytang/LLM-AggreFact")['test'])
 docs = df.doc.values
 claims = df.claim.values