All participants should upload their system description and output the TSV files. Participants are allowed to submit a total of 3 results (3 TSV files) per track. Please follow the submission format to be evaluated correctly. The formats are exactly the same as the ones used for the SLUE-Toolkit.

System description template

Overleaf link :

Submission format for pipeline track

In .tsv format, prepare 3 columns (id, pred_text, pred_ner) as in the below example. Please specify none, if the prediction does not exist.

id	pred_text	pred_ner
20150518-0900-PLENARY-15-en_20150518-18:48:27_2	we all agreed at the last session in strasbourg that development is important but we need to remember it now when we are talking about the financial contributions	[['GPE', 'strasbourg']]
20180530-0900-PLENARY-26-en_20180530-20:01:52_0	madam president first of all i would like to thank all the members who have participated in this important debate for their different contributions. None

Submission format for E2E track

in .tsv format, prepare 2 columns (id, pred_ner) as in the below example. pred_text column is optional depending on your system could generate ASR transcription. Please specify none, if the prediction does not exist.

id	pred_ner
20150518-0900-PLENARY-15-en_20150518-18:48:27_2	[['GPE', 'strasbourg']]
20180530-0900-PLENARY-26-en_20180530-20:01:52_0	 None