A Neural Language Model for Dynamically Representing the Meanings of Unknown Words and Entities in a Discourse
arXiv:1709.01679
Abstract
This study addresses the problem of identifying the meaning of unknown words or entities in a discourse with respect to the word embedding approaches used in neural language models. We proposed a method for on-the-fly construction and exploitation of word embeddings in both the input and output layers of a neural model by tracking contexts. This extends the dynamic entity representation used in Kobayashi et al. (2016) and incorporates a copy mechanism proposed independently by Gu et al. (2016) and Gulcehre et al. (2016). In addition, we construct a new task and dataset called Anonymized Language Modeling for evaluating the ability to capture word meanings while reading. Experiments conducted using our novel dataset show that the proposed variant of RNN language model outperformed the baseline model. Furthermore, the experiments also demonstrate that dynamic updates of an output layer help a model predict reappearing entities, whereas those of an input layer are effective to predict words following reappearing entities.
arXiv comment11 pages. To appear in IJCNLP 2017
References in corpus (12)
- Language Modeling with Gated Convolutional Networks
- Character-Aware Neural Language Models
- Exploring the Limits of Language Modeling
- Pointer Sentinel Mixture Models
- Incorporating Copying Mechanism in Sequence-to-Sequence Learning
- Tracking the World State with Recurrent Entity Networks
- Traversing Knowledge Graphs in Vector Space
- Learning to Compute Word Embeddings On the Fly
- Pointing the Unknown Words
- Learning Global Features for Coreference Resolution
- Linguistic Knowledge as Memory for Recurrent Neural Networks
- Two Discourse Driven Language Models for Semantics