Code and dataset supporting the paper:

Anusha Balakrishnan, Jinfeng Rao, Kartikeya Upasani, Michael White and Rajen Subba. Constrained Decoding for Neural NLG from Compositional Representations in Task-Oriented Dialogue. To appear in Proc. ACL-19.

If you find this code or dataset useful in your research, please consider citing our paper.

@InProceedings{Balakrishnan2019constrainednlg,
author = {Anusha Balakrishnan and Jinfeng Rao and Kartikeya Upasani and Michael White and Rajen Subba},
title = {Constrained Decoding for Neural {NLG} from Compositional Representations in Task-Oriented Dialogue},
booktitle = {Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics},
month = {July},
year = {2019},
note = {To appear}
}

In addition to the weather and enriched E2E challenge dataset from our paper, we released another weather_challenge dataset, which contains harder weather scenarios in train/val/test files. Each response was collected by providing annotators, who are native English speakers, with a user query, a compositional meaning representation (with discourse relations and dialog acts), and a context. Currently context is not provided yet but we will release soon. All of these are made available in our dataset. See our linked paper for more details.

Disc_Test is a more challenging subset of our test set that contains discourse relations, which is also the subset we report results in Disc column in Table 7 in our paper. Note that there are some minor differences of data statistics to our paper, please use the statistics above.

We noticed that slightly higher numbers can be obtained by tuning hyper-parameters compared to the numbers we reported in our paper. Therefore, we update all the automatic numbers (BLEU and tree accuracy) here and please use numbers below when citing our results. For tree accuracy, we report the number on the whole test set, as well as on two disjoint subsets: no-discourse subset that contains examples without any discourse act; discourse subset contains example with 1+ discourse acts.

Computing tree accuracy:

python compute_tree_acc.py -tsv ~/seq2seq_out.tsv

Output file should be tab-separated with columns id, input, pred, target.

TreeNLG is released under CC-BY-NC-4.0, see LICENSE for details.