Pointer + Multi-task Entailment and Paraphrase Generation

Reported on 6 benchmarks across 1 task · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing6 results

Text SimplificationonTurkCorpus
SARI (EASSE>=0.2.1)· 2018-06-19
37.45
best: 43.46 (GPT-175B (6 SARI-selected examples, high/low))
SOTA
Dynamic Multi-Level Multi-Task Learning for Sentence Simplification arXiv:1806.07304
Text SimplificationonNewsela
SARI· 2018-06-19
33.22
best: 36.6 (CRF Alignment + Transformer)
SOTA
Dynamic Multi-Level Multi-Task Learning for Sentence Simplification arXiv:1806.07304
Text SimplificationonTurkCorpus
BLEU· 2018-06-19
81.49
best: 92.02 (NSELSTM-B)
Dynamic Multi-Level Multi-Task Learning for Sentence Simplification arXiv:1806.07304
Text SimplificationonNewsela
BLEU· 2018-06-19
11.14
best: 26.31 (NSELSTM-B )
Dynamic Multi-Level Multi-Task Learning for Sentence Simplification arXiv:1806.07304
Text SimplificationonPWKP / WikiSmall
BLEU· 2018-06-19
27.23
best: 53.42 (NSELSTM-B)
Dynamic Multi-Level Multi-Task Learning for Sentence Simplification arXiv:1806.07304
Text SimplificationonPWKP / WikiSmall
SARI· 2018-06-19
29.58
best: 44.67 (TST)
Dynamic Multi-Level Multi-Task Learning for Sentence Simplification arXiv:1806.07304