Skip to content
Tensorflow Source code for "Recurrently Controlled Recurrent Networks" (NIPS 2018)
Branch: master
Clone or download

Latest commit

Fetching latest commit…
Cannot retrieve the latest commit at this time.


Type Name Latest commit message Commit time
Failed to load latest commit information.

NIPS 2018

We propose a new Recurrently controlled recurrent networks (RCRN) that shows some improvement over stacked BiLSTMs and BiLSTM across a number of NLP tasks.

This repository contains the Tensorflow model file for RCRN, according with the custom cuda optimized kernel. I will upload running scripts / example usage when I have time (already have tons of backlog ): ) .


Python 2.7 Tensorflow 1.7


Our CUDA op was adapted from:

Cudnn RNN was adapted from:


If you find our repository useful, please cite our paper!

  author    = {Yi Tay and
               Luu Anh Tuan and
               Siu Cheung Hui},
  title     = {Recurrently Controlled Recurrent Networks},
  booktitle = {Proceedings of NIPS 2018},
  year      = {2018}
You can’t perform that action at this time.