@inproceedings{petroni2019language, title={Language Models as Knowledge Bases?}, author={Fabio Petroni and Tim Rocktäschel and Patrick Lewis and Anton Bakhtin and Yuxiang Wu and Alexander H. Miller and Sebastian Riedel}, url="https://arxiv.org/abs/1909.01066" year = "2019", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = nov, publisher = "Association for Computational Linguistics" }

@inproceedings{lewis2019unsupervisedqa, title = "Unsupervised Question Answering by Cloze Translation author = "Lewis, Patrick and Denoyer, Ludovic and Riedel, Sebastian", booktitle = "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", month = aug, year = "2019", address = "Florence, Italy", publisher = "Association for Computational Linguistics", url = "https://arxiv.org/abs/1906.04980", abstract="Obtaining training data for Question Answering (QA) is time-consuming and resource-intensive, and existing QA datasets are only available for limited domains and languages. In this work, we explore to what extent high quality training data is actually required for Extractive QA, and investigate the possibility of unsupervised Extractive QA. We approach this problem by first learning to generate context, question and answer triples in an unsupervised manner, which we then use to synthesize Extractive QA training data automatically. To generate such triples, we first sample random context paragraphs from a large corpus of documents and then random noun phrases or named entity mentions from these paragraphs as answers. Next we convert answers in context to "fill-in-the-blank" cloze questions and finally translate them into natural questions. We propose and compare various unsupervised ways to perform cloze-to-natural question translation, including training an unsupervised NMT model using non-aligned corpora of natural questions and cloze questions as well as a rule-based approach. We find that modern QA models can learn to answer human questions surprisingly well using only synthetic training data. We demonstrate that, without using the SQuAD training data at all, our approach achieves 56.4 F1 on SQuAD v1 (64.5 F1 when the answer is a Named entity mention), outperforming early supervised models." }

@inproceedings{hosking-riedel-2019-evaluating, title = "Evaluating Rewards for Question Generation Models", author = "Hosking, Tom and Riedel, Sebastian", booktitle = "Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)", month = jun, year = "2019", address = "Minneapolis, Minnesota", publisher = "Association for Computational Linguistics", url = "https://www.aclweb.org/anthology/N19-1237", pages = "2278--2283", abstract = "Recent approaches to question generation have used modifications to a Seq2Seq architecture inspired by advances in machine translation. Models are trained using teacher forcing to optimise only the one-step-ahead prediction. However, at test time, the model is asked to generate a whole sequence, causing errors to propagate through the generation process (exposure bias). A number of authors have suggested that optimising for rewards less tightly coupled to the training data might counter this mismatch. We therefore optimise directly for various objectives beyond simply replicating the ground truth questions, including a novel approach using an adversarial discriminator that seeks to generate questions that are indistinguishable from real examples. We confirm that training with policy gradient methods leads to increases in the metrics used as rewards. We perform a human evaluation, and show that although these metrics have previously been assumed to be good proxies for question quality, they are poorly aligned with human judgement and the model simply learns to exploit the weaknesses of the reward source.", }

@article{DBLP:journals/corr/abs-1809-01494, author = {Marzieh Saeidi and Max Bartolo and Patrick Lewis and Sameer Singh and Tim Rockt{\"{a}}schel and Mike Sheldon and Guillaume Bouchard and Sebastian Riedel}, title = {Interpretation of Natural Language Rules in Conversational Machine Reading}, journal = {CoRR}, volume = {abs/1809.01494}, year = {2018}, url = {http://arxiv.org/abs/1809.01494}, archivePrefix = {arXiv}, eprint = {1809.01494}, timestamp = {Fri, 05 Oct 2018 11:34:52 +0200}, biburl = {https://dblp.org/rec/bib/journals/corr/abs-1809-01494}, bibsource = {dblp computer science bibliography, https://dblp.org} }

@article{welbl2018tacl, volume={6}, journal={Transactions of the Association for Computational Linguistics}, pages={287--302}, keywords={}, issn={2307-387X}, year={2018}, author={Johannes Welbl, Pontus Stenetorp, Sebastian Riedel}, title={Constructing Datasets for Multi-hop Reading Comprehension Across Documents}, }