ARTE is a manually constructed corpus of semantic-linguistically annotated textual entailment data from the Second PASCAL Recognizing Textual Entailment (RTE-2) Challenge.
The annotation format is the one of the open-source MMAX2 annotation tool, which can be used for visualizing and accessing the data. The annotation scheme, process and guidelines are fully described here:
Konstantina Garoufi. Towards a better understanding of applied textual entailment: Annotation and evaluation of the RTE-2 dataset. Master’s thesis, Saarland University, 2007.
For further information or questions, please contact me.