Update README.md

README.md

@@ -32,43 +32,6 @@ task_categories:  # Full list at https://github.com/huggingface/hub-docs/blob/ma
 task_ids:
 - multi-class-classification
 paperswithcode_id: multitacred
-configs:  # Optional for datasets with multiple configurations like glue.
-- original-ar
-- original-de
-- original-es
-- original-fi
-- original-fr
-- original-hi
-- original-hu
-- original-ja
-- original-pl
-- original-ru
-- original-tr
-- original-zh
-- revisited-ar
-- revisited-de
-- revisited-es
-- revisited-fi
-- revisited-fr
-- revisited-hi
-- revisited-hu
-- revisited-ja
-- revisited-pl
-- revisited-ru
-- revisited-tr
-- revisited-zh
-- retacred-ar
-- retacred-de
-- retacred-es
-- retacred-fi
-- retacred-fr
-- retacred-hi
-- retacred-hu
-- retacred-ja
-- retacred-pl
-- retacred-ru
-- retacred-tr
-- retacred-zh
 dataset_info:
 - config_name: original-ar
   features:
@@ -4865,12 +4828,16 @@ subject and object entity markup still intact, these were discarded.
 
 ## Dataset Creation
 ### Curation Rationale
-[More Information Needed]
+To enable more research on multilingual Relation Extraction, we generate translations of the TAC relation extraction
+dataset using DeepL and Google Translate.
 ### Source Data
 #### Initial Data Collection and Normalization
-[More Information Needed]
+The instances of this dataset are sentences from the
+[original TACRED dataset](https://nlp.stanford.edu/projects/tacred/), which in turn
+are sampled from the [corpus](https://catalog.ldc.upenn.edu/LDC2018T03) used in the yearly 
+[TAC Knowledge Base Population (TAC KBP) challenges](https://tac.nist.gov/2017/KBP/index.html).
 #### Who are the source language producers?
-[More Information Needed]
+Newswire and web texts collected for the [TAC Knowledge Base Population (TAC KBP) challenges](https://tac.nist.gov/2017/KBP/index.html).
 ### Annotations
 #### Annotation process
 See the Stanford paper, the TACRED Revisited paper, and the Re-TACRED paper, plus their appendices, for
@@ -4879,19 +4846,24 @@ details on the original annotation process. The translated versions do not chang
 Translations were tokenized with language-specific Spacy models (Spacy 3.1, 'core_news/web_sm' models)
 or Trankit (Trankit 1.1.0) when there was no Spacy model for a given language (Hungarian, Turkish, Arabic, Hindi).
 #### Who are the annotators?
-[More Information Needed]
+The original TACRED dataset was annotated by crowd workers, see the [TACRED paper](https://nlp.stanford.edu/pubs/zhang2017tacred.pdf).
 ### Personal and Sensitive Information
-[More Information Needed]
+The [authors](https://nlp.stanford.edu/pubs/zhang2017tacred.pdf) of the original TACRED dataset
+have not stated measures that prevent collecting sensitive or offensive text. Therefore, we do 
+not rule out the possible risk of sensitive/offensive content in the translated data.
 ## Considerations for Using the Data
 ### Social Impact of Dataset
-[More Information Needed]
+not applicable
 ### Discussion of Biases
-[More Information Needed]
+The dataset is drawn from web and newswire text, and thus reflects any biases of these original
+texts, as well as biases introduced by the MT models.
 ### Other Known Limitations
-[More Information Needed]
+not applicable
 ## Additional Information
 ### Dataset Curators
-[More Information Needed]
+The dataset was created by members of the
+[DFKI SLT team: Leonhard Hennig, Philippe Thomas, Sebastian Möller, Gabriel Kressin](https://www.dfki.de/en/web/research/research-departments/speech-and-language-technology/speech-and-language-technology-staff-members)
+
 ### Licensing Information
 To respect the copyright of the underlying TACRED dataset, MultiTACRED is released via the
 Linguistic Data Consortium ([LDC License](https://catalog.ldc.upenn.edu/license/ldc-non-members-agreement.pdf)).