Update README.md

README.md

@@ -165,7 +165,14 @@ license_link: https://creativecommons.org/licenses/by-nc/4.0/
 It is a test set with single and multi-page invoices sourced from the Federal Communications Commission (FCC) to evaluate key information extraction (KIE) performance. 
 
 ### Task
-Extract information from the document in JSON format given the corresponding JSON schema. It contains 75 documents, with: a) a common JSON schema and b) corresponding expected JSON responses. 
+Extract information from the document in JSON format given the corresponding JSON schema. It contains 75 documents, with: 
+
+a) `image_files`: Each document has multiple pages
+
+b) `json_schema`: A common JSON schema requiring extraction of specified information including line items in the invoices
+
+c) `json_response`: This is the expected output from the model
+
 
 ### Modifications to the original dataset
 This release resolves two issues present in the original evaluation set (https://arxiv.org/abs/2403.20101):
@@ -174,6 +181,10 @@ a) Line Item Grouping: Fields previously treated as independent entries are now
 
 b) Annotation Corrections: Erroneous values in the original annotations have been corrected to ensure data accuracy.
 
+These corrections are applied to `json_schema` and `json_response`.
+
+Note: We haven't edited the OCR (`text`) field, and instead use the original version.
+
 ### License
 ```
 Document Source: FCC https://publicfiles.fcc.gov/