Analyzing Bayesian Crosslingual Transfer in Topic Models.
2019 CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: HUMAN LANGUAGE TECHNOLOGIES (NAACL HLT 2019), VOL. 1(2019)
摘要
S, T Source and target languages. They are interchangeable during Gibbs sampling. For example, when training English and German, English can be either source or target. w` A word type of language `. x` An individual token of language `. zx` The topic assignment of token x`. Sw` The sample of word typew`, the set containing all the tokens x` that are of this word type. Px` , Px`,k Px` denotes the conditional distribution over all topics for token x`. The conditional probability of sampling a topic k from Px` is denoted as Px`,k. D(`) The set of documents in language `. This usually refers to the test corpus. D̂(`) The array of document representations from the corpus D(`) and their document labels.
更多查看译文
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要