Zheng Huang

Bi-directional LSTM Recurrent Neural Network for Chinese Word Segmentation

2016 · arXiv (Cornell University)

Recurrent neural network(RNN) has been broadly applied to natural language processing(NLP) problems. This kind of neural network is designed for modeling sequential data and has been testified to be quite efficient in sequential tagging tasks. …

Lensless Opto-Electronic Neural Network Architecture For Processing Multi-Color-Channel Signals

2023

The lensless opto-electronic neural network based on a multi-channel mask enables completing the convolution operations of multi-color-channel signals on the light field, significantly improving the task accuracy, eliminating the sensor components, and reducing energy consumption.

Generalizable Retrieve-Based Method for Task Oriented Dialogue State Tracking

2025

Dialogue state tracking(DST) is an essential component of task oriented dialogue(TOD) system. Recent research in schema guided DST utilizing semantic information offered by schema has seen great progress in transferring to new unseen domains. However, …