NAACL 2018 Papers

We propose a simple but highly effective automatic evaluation measure of summarization, pruned Basic Elements (pBE). Although the BE concept is widely used for the automated evaluation of summaries, its weakness is that it redundantly matches basic elements. To avoid this redundancy, pBE prunes basic elements by (1) disregarding frequency count of basic elements and (2) reducing semantically overlapped basic elements based on word similarity. Even though it is simple, pBE outperforms ROUGE in DUC datasets in most cases and achieves the highest rank correlation coefficient in TAC 2011 AESOP task.

Supervised summarization systems usually rely on supervision at the sentence or n-gram level provided by automatic metrics like ROUGE, which act as noisy proxies for human judgments. In this work, we learn a summary-level scoring function \theta including human judgments as supervision and automatically generated data as regularization. We extract summaries with a genetic algorithm using \theta as a fitness function. We observe strong and promising performances across datasets in both automatic and manual evaluation.

Abstractive text summarization is the task of compressing and rewriting a long document into a short summary while maintaining saliency, directed logical entailment, and non-redundancy. In this work, we address these three important aspects of a good summary via a reinforcement learning approach with two novel reward functions: ROUGESal and Entail, on top of a coverage-based baseline. The ROUGESal reward modifies the ROUGE metric by up-weighting the salient phrases/words detected via a keyphrase classifier. The Entail reward gives high (length-normalized) scores to logically-entailed summaries using an entailment classifier. Further, we show superior performance improvement when these rewards are combined with traditional metric (ROUGE) based rewards, via our novel and effective multi-reward approach of optimizing multiple rewards simultaneously in alternate mini-batches. Our method achieves the new state-of-the-art results on CNN/Daily Mail dataset as well as strong improvements in a test-only transfer setup on DUC-2002.

While Wikipedia exists in 287 languages, its content is unevenly distributed among them. In this work, we investigate the generation of open domain Wikipedia summaries in underserved languages using structured data from Wikidata. To this end, we propose a neural network architecture equipped with copy actions that learns to generate single-sentence and comprehensible textual summaries from Wikidata triples. We demonstrate the effectiveness of the proposed approach by evaluating it against a set of baselines on two languages of different natures: Arabic, a morphological rich language with a larger vocabulary than English, and Esperanto, a constructed language known for its easy acquisition.

Keyphrase extraction is a fundamental task in natural language processing that facilitates mapping of documents to a set of representative phrases. In this paper, we present an unsupervised technique (Key2Vec) that leverages phrase embeddings for ranking keyphrases extracted from scientific articles. Specifically, we propose an effective way of processing text documents for training multi-word phrase embeddings that are used for thematic representation of scientific articles and ranking of keyphrases extracted from them using theme-weighted PageRank. Evaluations are performed on benchmark datasets producing state-of-the-art results.

Most summarization research focuses on summarizing the entire given text, but in practice readers are often interested in only one aspect of the document or conversation. We propose targeted summarization as an umbrella category for summarization tasks that intentionally consider only parts of the input data. This covers query-based summarization, update summarization, and a new task we propose where the goal is to summarize a particular aspect of a document. However, collecting data for this new task is hard because directly asking annotators (e.g., crowd workers) to write summaries leads to data with low accuracy when there are a large number of facts to include. We introduce a novel crowdsourcing workflow, Pin-Refine, that allows us to collect high-quality summaries for our task, a necessary step for the development of automatic systems.

Structured data summarization involves generation of natural language summaries from structured input data. In this work, we consider summarizing structured data occurring in the form of tables as they are prevalent across a wide variety of domains. We formulate the standard table summarization problem, which deals with tables conforming to a single predefined schema. To this end, we propose a mixed hierarchical attention based encoder-decoder model which is able to leverage the structure in addition to the content of the tables. Our experiments on the publicly available weathergov dataset show around 18 BLEU (around 30%) improvement over the current state-of-the-art.

Neural abstractive summarization models have led to promising results in summarizing relatively short documents. We propose the first model for abstractive summarization of single, longer-form documents (e.g., research papers). Our approach consists of a new hierarchical encoder that models the discourse structure of a document, and an attentive discourse-aware decoder to generate the summary. Empirical results on two large-scale datasets of scientific papers show that our model significantly outperforms state-of-the-art models.

Social media is known for its multi-cultural and multilingual interactions, a natural product of which is code-mixing. Multilingual speakers mix languages they tweet to address a different audience, express certain feelings, or attract attention. This paper presents a large-scale analysis of 6 million tweets produced by 27 thousand multilingual users speaking 12 other languages besides English. We rely on this corpus to build predictive models to infer non-English languages that users speak exclusively from their English tweets. Unlike native language identification task, we rely on large amounts of informal social media communications rather than ESL essays. We contrast the predictive power of the state-of-the-art machine learning models trained on lexical, syntactic, and stylistic signals with neural network models learned from word, character and byte representations extracted from English only tweets. We report that content, style and syntax are the most predictive of non-English languages that users speak on Twitter. Neural network models learned from byte representations of user content combined with transfer learning yield the best performance. Finally, by analyzing cross-lingual transfer – the influence of non-English languages on various levels of linguistic performance in English, we present novel findings on stylistic and syntactic variations across speakers of 12 languages in social media.

With the growing amount of reviews in e-commerce websites, it is critical to assess the helpfulness of reviews and recommend them accordingly to consumers. Recent studies on review helpfulness require plenty of labeled samples for each domain/category of interests. However, such an approach based on close-world assumption is not always practical, especially for domains with limited reviews or the “out-of-vocabulary” problem. Therefore, we propose a convolutional neural network (CNN) based model which leverages both word-level and character-based representations. To transfer knowledge between domains, we further extend our model to jointly model different domains with auxiliary domain discriminators. On the Amazon product review dataset, our approach significantly outperforms the state of the art in terms of both accuracy and cross-domain robustness.

This paper addresses the problem of community membership detection using only text features in a scenario where a small number of positive labeled examples defines the community. The solution introduces an unsupervised proxy task for learning user embeddings: user re-identification. Experiments with 16 different communities show that the resulting embeddings are more effective for community membership identification than common unsupervised representations.

Interacting with relational databases through natural language helps users with any background easily query and analyze a vast amount of data. This requires a system that understands users’ questions and converts them to SQL queries automatically. In this paper, we present a novel approach TypeSQL which formats the problem as a slot filling task in a more reasonable way. In addition, TypeSQL utilizes type information to better understand rare entities and numbers in the questions. We experiment this idea on the WikiSQL dataset and outperform the prior art by 6% in much shorter time. We also show that accessing the content of databases can significantly improve the performance when users’ queries are not well-formed. TypeSQL can reach 82.6% accuracy, a 17.5% absolute improvement compared to the previous content-sensitive model.

Recent success of deep learning models for the task of extractive Question Answering (QA) is hinged on the availability of large annotated corpora. However, large domain specific annotated corpora are limited and expensive to construct. In this work, we envision a system where the end user specifies a set of base documents and only a few labelled examples. Our system exploits the document structure to create cloze-style questions from these base documents; pre-trains a powerful neural network on the cloze style questions; and further fine-tunes the model on the labeled examples. We evaluate our proposed system across three diverse datasets from different domains, and find it to be highly effective with very little labeled data. We attain more than 50% F1 score on SQuAD and TriviaQA with less than a thousand labelled examples. We are also releasing a set of 3.2M cloze-style questions for practitioners to use while building QA systems.

It is shown that many published models for the Stanford Question Answering Dataset (Rajpurkar et al., 2016) lack robustness, suffering an over 50% decrease in F1 score during adversarial evaluation based on the AddSent (Jia and Liang, 2017) algorithm. It has also been shown that retraining models on data generated by AddSent has limited effect on their robustness. We propose a novel alternative adversary-generation algorithm, AddSentDiverse, that significantly increases the variance within the adversarial training data by providing effective examples that punish the model for making certain superficial assumptions. Further, in order to improve robustness to AddSent’s semantic perturbations (e.g., antonyms), we jointly improve the model’s semantic-relationship learning capabilities in addition to our AddSentDiverse-based adversarial training data augmentation. With these additions, we show that we can make a state-of-the-art model significantly more robust, achieving a 36.5% increase in F1 score under many different types of adversarial evaluation while maintaining performance on the regular SQuAD task.

The task of natural question generation is to generate a corresponding question given the input passage (fact) and answer. It is useful for enlarging the training set of QA systems. Previous work has adopted sequence-to-sequence models that take a passage with an additional bit to indicate answer position as input. However, they do not explicitly model the information between answer and other context within the passage. We propose a model that matches the answer with the passage before generating the question. Experiments show that our model outperforms the existing state of the art using rich features.

We introduce Question-Answer Meaning Representations (QAMRs), which represent the predicate-argument structure of a sentence as a set of question-answer pairs. We develop a crowdsourcing scheme to show that QAMRs can be labeled with very little training, and gather a dataset with over 5,000 sentences and 100,000 questions. A qualitative analysis demonstrates that the crowd-generated question-answer pairs cover the vast majority of predicate-argument relationships in existing datasets (including PropBank, NomBank, and QA-SRL) along with many previously under-resourced ones, including implicit arguments and relations. We also report baseline models for question generation and answering, and summarize a recent approach for using QAMR labels to improve an Open IE system. These results suggest the freely available QAMR data and annotation scheme should support significant future work.

Reading a document and extracting an answer to a question about its content has attracted substantial attention recently. While most work has focused on the interaction between the question and the document, in this work we evaluate the importance of context when the question and document are processed independently. We take a standard neural architecture for this task, and show that by providing rich contextualized word representations from a large pre-trained language model as well as allowing the model to choose between context-dependent and context-independent word representations, we can obtain dramatic improvements and reach performance comparable to state-of-the-art on the competitive SQuAD dataset.

Lexical ambiguity makes it difficult to compute useful statistics of a corpus. A given word form might represent any of several morphological feature bundles. One can, however, use unsupervised learning (as in EM) to fit a model that probabilistically disambiguates word forms. We present such an approach, which employs a neural network to smoothly model a prior distribution over feature bundles (even rare ones). Although this basic model does not consider a token’s context, that very property allows it to operate on a simple list of unigram type counts, partitioning each count among different analyses of that unigram. We discuss evaluation metrics for this novel task and report results on 5 languages.

We analyze the complexity of the problem of determining whether a set of phonemes forms a natural class and, if so, that of finding the minimal feature specification for the class. A standard assumption in phonology is that finding a minimal feature specification is an automatic part of acquisition and generalization. We find that the natural class decision problem is tractable (i.e. is in P), while the minimization problem is not; the decision version of the problem which determines whether a natural class can be defined with k features or less is NP-complete. We also show that, empirically, a greedy algorithm for finding minimal feature specifications will sometimes fail, and thus cannot be assumed to be the basis for human performance in solving the problem.