分类
字节跳动 | AI Lab | NLP算法工程师
论文赏析[TACL19]生成模型还在用自左向右的顺序?这篇论文教你如何自动推测最佳生成顺序 论文赏析[TACL19]生成模型还在用自左向右的顺序?这篇论文教你如何自动推测最佳生成顺序
关注公众号【算法码上来】,每日算法干货马上就来! 论文地址:Insertion-based Decoding with automatically Inferred Generation Order 介绍大多数的生成模型(例如seq2
2020-01-05
论文赏析[ACL18]一个句子向量表示究竟可以塞进多少语言性质? 论文赏析[ACL18]一个句子向量表示究竟可以塞进多少语言性质?
关注公众号【算法码上来】,每日算法干货马上就来! 论文地址:What you can cram into a single $&!#* vector: Probing sentence embeddings for lingu
2019-12-25
论文赏析[EMNLP19]Designing and Interpreting Probes with Control Tasks 论文赏析[EMNLP19]Designing and Interpreting Probes with Control Tasks
关注公众号【算法码上来】,每日算法干货马上就来! 论文地址:Designing and Interpreting Probes with Control Tasks 介绍NLP任务都存在着一个常见的问题:我的神经网络有没有学到某个性质
2019-12-19
成分句法分析综述(第二版) 成分句法分析综述(第二版)
本文对成分句法分析近年来的进展做了一个比较完善的总结。分析了多种不同类型的成分句法分析模型(基于转移,动态规划和序列到序列等),比较了它们之间的优缺点,并总结了一些提升它们性能的技巧。最后,本文对成分句法分析的未来发展趋势表明了自己的一些看法。
2019-08-15
Do latent tree learning models identify meaningful structure in sentences? Do latent tree learning models identify meaningful structure in sentences?
关注公众号【算法码上来】,每日算法干货马上就来! 论文地址:Do latent tree learning models identify meaningful structure in sentences? 本文是一篇分析类论文,主
2019-08-05
Unsupervised Latent Tree Induction with Deep Inside-Outside Recursive Autoencoders Unsupervised Latent Tree Induction with Deep Inside-Outside Recursive Autoencoders
关注公众号【算法码上来】,每日算法干货马上就来! 论文地址:Unsupervised Latent Tree Induction with Deep Inside-Outside Recursive Autoencoders代码地址:
2019-07-25
Unsupervised Recurrent Neural Network Grammars Unsupervised Recurrent Neural Network Grammars
关注公众号【算法码上来】,每日算法干货马上就来! 论文地址:Unsupervised Recurrent Neural Network Grammars代码地址:github 介绍 这篇是新鲜出炉的NAACL19的关于无监督循环神经网
2019-04-20
Neural Language Modeling by Jointly Learning Syntax and Lexicon Neural Language Modeling by Jointly Learning Syntax and Lexicon
关注公众号【算法码上来】,每日算法干货马上就来! 论文地址:Neural Language Modeling by Jointly Learning Syntax and Lexicon代码地址:github 最近开始转向去看看一些
2019-03-31
Better, Faster, Stronger Sequence Tagging Constituent Parsers Better, Faster, Stronger Sequence Tagging Constituent Parsers
关注公众号【算法码上来】,每日算法干货马上就来! 为了看懂论文里的策略梯度,又去把强化学习看了一遍。。。 论文地址:Better, Faster, Stronger Sequence Tagging Constituent Par
2019-03-11
Constituent Parsing as Sequence Labeling Constituent Parsing as Sequence Labeling
关注公众号【算法码上来】,每日算法干货马上就来! 貌似已经有好几个月没怎么看过论文了,之前一直在写论文,一直没空更新博客,最近闲下来把最后几篇没看完的论文看了。 论文地址:Constituent Parsing as Sequen
2019-03-11
Dynamic Oracles for Top-Down and In-Order Shift-Reduce Constituent Parsing Dynamic Oracles for Top-Down and In-Order Shift-Reduce Constituent Parsing
关注公众号【算法码上来】,每日算法干货马上就来! 有一句话,宾语是你。“吉下两点一口,又有欠字相依。” 论文地址:Dynamic Oracles for Top-Down and In-Order Shift-Reduce Con
2018-11-07
Faster Shift-Reduce Constituent Parsing with a Non-Binary, Bottom-Up Strategy Faster Shift-Reduce Constituent Parsing with a Non-Binary, Bottom-Up Strategy
关注公众号【算法码上来】,每日算法干货马上就来! 论文地址:Faster Shift-Reduce Constituent Parsing with a Non-Binary, Bottom-Up Strategy 介绍 这篇论文提出
2018-11-06
2 / 5