DMV + invertible projector

Reported on 2 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 Processing2 results

Constituency ParsingonPTB Diagnostic ECG Database
Mean F1 (WSJ)· 2018-08-28
47.9
best: 70.4 (Ensemble (Generative MBR))
SOTA
Unsupervised Learning of Syntactic Structure with Invertible Neural Projections arXiv:1808.09111
Constituency ParsingonPTB Diagnostic ECG Database
Mean F1 (WSJ10)· 2018-08-28
60.2
best: 74.2 (inside-outside co-training + weak supervision)
SOTA
Unsupervised Learning of Syntactic Structure with Invertible Neural Projections arXiv:1808.09111