id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_4zamv3kd6bhadgzzpkitxfvryi	Waleed Ammar	Many Languages, One Parser	2016	14	.pdf	application/pdf	8534	855	64	uses (i) multilingual word clusters and embeddings; (ii) token-level language information; and (iii) language-specific features (finegrained POS tags). POS tags, multilingual word embeddings and multilingual word clusters, and (2) tweaking the behavior of the parser depending on the current input Therefore, we extend the token representation in MALOPA by concatenating learned embeddings of multilingual word clusters, and pretrained 'MultiCCA' uses a linear operator to project pretrained monolingual embeddings in each language (except English) to the For tagging, we construct the token representation by concatenating the embeddings of the word type (pretrained), the Brown cluster and the input language. The token representation used for parsing includes the embedding of predicted POS tags, which may be incorrect. Table 3: Dependency parsing: labeled attachment scores (LAS) for monolingually-trained parsers and fine-grained POS embeddings, on average outperforms monolingually-trained parsers for target languages with a treebank.	./cache/work_4zamv3kd6bhadgzzpkitxfvryi.pdf	./txt/work_4zamv3kd6bhadgzzpkitxfvryi.txt