Filtered AMR Graphs Bank

The filtered AMR graphs bank is the output of the Polifonia Knowledge Extractor pipeline. The Polifonia Mini Textual Corpus is pre-processed through coreference resolution and minimal rule-based post-OCR correction, then it is given as input to SPRING to obtain AMR graphs. The filtered AMR graphs bank encompasses the sentence-AMR graph pairs extrapolated from our initial AMR graphs bank corresponding to AMR2Text-generated sentences associated with a positive (>0) BLEURT score. Full description at: https://github.com/polifonia-project/Polifonia-Knowledge-Extractor