The code implements the neural encoder-aligner-decoder model for direction following task, described in paper "Listen, Attend, and Walk: Neural Mapping of Navigational Instructions to Action Sequences" (Press the link http://arxiv.org/abs/1506.04089). The paper will appear at AAAI-2016 (Oral) and also at NIPS-2015 Multimodal Machine Learning workshop (Oral).
The code is under final-testing and will show up soon...