Unsupervised Latent Tree Induction with Deep Inside-Outside Recursive Auto-Encoders
Open paper
Benchmark
Model
Rank
Results
constituency-grammar-induction-on-ptb
DIORA (+PP)
#14
Mean F1 (WSJ): 55.7
Max F1 (WSJ): 56.2
Max F1 (WSJ10): 60.55
constituency-grammar-induction-on-ptb
DIORA
#18
Mean F1 (WSJ): 48.9
Max F1 (WSJ): 49.6
Mean F1 (WSJ10): 67.7
…
Rank counts only results with a code link.