Gated Word-Character Recurrent Language Model

2016/06/06 by Yasumasa Miyamoto, Kyunghyun Cho, Miyamoto, Yasumasa +1 · 2 citations
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Speech Recognition and Synthesis #Topic Modeling

paper · pdf · doi:10.48550/arxiv.1606.01700

openalex publication_date 2016/06/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We introduce a recurrent neural network language model (RNN-LM) with long short-term memory (LSTM) units that utilizes both character-level and word-level inputs. Our model has a gate that adaptively finds the optimal mixture of the character-level and word-level inputs. The gate creates the final vector representation of a word by combining two distinct representations of the word. The character-level inputs are converted into vector representations of words using a bidirectional LSTM. The word-level inputs are projected into another high-dimensional space by a word lookup table. The final vector representations of words are used in the LSTM language model which predicts the next word given all the preceding words. Our model with the gating mechanism effectively utilizes the character-level inputs for rare and out-of-vocabulary words and outperforms word-level language models on several English corpora.

Citations

• Recurrent Neural Network Regularization
• Exploring the Limits of Language Modeling
• Finding Function in Form: Compositional Character Models for Open Vocabulary Word Representation

Cited by

• Trellis Networks for Sequence Modeling
• An Embedded Deep Learning based Word Prediction

Related

• Character-Aware Neural Language Models
• Character-Level Language Modeling with Hierarchical Recurrent Neural Networks
• Alternative structures for character-level RNNs
• Language Models with Pre-Trained (GloVe) Word Embeddings
• Character n-gram Embeddings to Improve RNN Language Models