CaRB
Crowdsourced automatic open Relation extraction Benchmark
Introduced 2019-11-01
CaRB [Bhardwaj et al., 2019] is developed by re-annotating the dev and test splits of OIE2016 via crowd-sourcing. Besides improving annotation quality, CaRB also provides a new matching scorer. CaRB scorer uses token level match and it matches relation with relation, arguments with arguments.
Source: https://arxiv.org/pdf/2205.11725.pdf (section 3.1)