WiRe57
Introduced by Léchelle et al. in WiRe57 : A Fine-Grained Benchmark for Open Information Extraction
We manually performed the task of Open Information Extraction on 5 short documents, elaborating tentative guidelines for the task, and resulting in a ground truth reference of 347 tuples. [section 1]
A small corpus of 57 sentences taken from the beginning of 5 documents in English was used as the source text from which to extract tuples. Three documents are Wikipedia articles (Chilly Gonzales, the EM algorithm, and Tokyo) and two are newswire articles (taken from Reuters, hence the Wi-Re name). [section 3.1]
Benchmarks
| Trend | Task | Dataset Variant | Best Model | Paper | Code |
|---|---|---|---|---|---|
|
WiRe57
|
CIGL-OIE + IGL-CA
|
Papers
| Paper | Code | Results | Date | Stars |
|---|
Dataset Loaders
No data loaders found. You can
submit your data loader here.
No data loaders found. You can
submit your data loader here.
