diff --git a/semeval2015_data/README b/semeval2015_data/README
index b7200d2..8cc05b9 100644
--- a/semeval2015_data/README
+++ b/semeval2015_data/README
@@ -1,37 +1,42 @@
-To reproduce the SemEval 2014 paper [1]:
+To reproduce the SemEval 2015 paper [1]:
 
-1) Place the SemEval 2014 original data in folders "train" and "test".
+1) Place the SemEval 2015 original data in folders "train" and "test".
 The directory contents should be:
 
-train/sb.bn.cpn
-train/dm.sdp
-train/pas.sdp
-train/pcedt.sdp
-train/sb.berkeley.cpn
-
-test/dm.sdp
-test/pas.sdp
-test/pcedt.sdp
-test/sb.berkeley.cpn
-test/sb.bn.cpn
+train/cs.psd.sdp
+train/cz.pas.sdp
+train/en.dm.sdp
+train/en.pas.sdp
+train/en.psd.sdp
+train/en.sb.bn.cpn
+
+test/cs.id.psd.sdp
+test/cs.ood.psd.sdp
+test/cz.id.pas.sdp
+test/en.id.dm.sdp
+test/en.id.pas.sdp
+test/en.id.psd.sdp
+test/en.id.sb.bn.cpn
+test/en.ood.dm.sdp
+test/en.ood.pas.sdp
+test/en.ood.psd.sdp
+test/en.ood.sb.bn.cpn
 
 2) Type:
 
 cd scripts
 ./generate_all_splits.sh
 
-This will create folders "dm", "pas", "pcedt" where the train/dev/test splits
+This will create folders "dm", "pas", "psd" where the train/dev/test splits
 will be stored.
 
 3) Type:
 
 cd ../../scripts_srl
-./train_test_submission_closed.sh
-./train_test_submission_open.sh
+./train_test_submission_SDP2015.sh
 
 It should take a few hours to train and run the semantic parsers.
 
-
-[1] André F.T. Martins and Mariana S.C. Almeida (2014).
-"Priberam: A Turbo Semantic Parser with Second Order Features."
-Proc. of SemEval 2014 task 8 (Broad-Coverage Semantic Dependency Parsing).
+[1] Mariana S.C. Almeida and André F.T. Martins (2015).
+"Lisbon: Evaluating TurboSemanticParser on Multiple Languages and Out-of-Domain Data."
+Proc. of SemEval 2015 task 18 (Broad-Coverage Semantic Dependency Parsing).