EMNLP 2018 Papers

We propose a model for tagging unstructured texts with an arbitrary number of terms drawn from a tree-structured vocabulary (i.e., an ontology). We treat this as a special case of sequence-to-sequence learning in which the decoder begins at the root node of an ontological tree and recursively elects to expand child nodes as a function of the input text, the current node, and the latent decoder state. We demonstrate that this method yields state-of-the-art results on the important task of assigning MeSH terms to biomedical abstracts.

Adversarial training (AT) is a regularization method that can be used to improve the robustness of neural network methods by adding small perturbations in the training data. We show how to use AT for the tasks of entity recognition and relation extraction. In particular, we demonstrate that applying AT to a general purpose baseline model for jointly extracting entities and relations, allows improving the state-of-the-art effectiveness on several datasets in different contexts (i.e., news, biomedical, and real estate data) and for different languages (English and Dutch).

Extracting typed entity mentions from text is a fundamental component to language understanding and reasoning. While there exist substantial labeled text datasets for multiple subsets of biomedical entity types—such as genes and proteins, or chemicals and diseases—it is rare to find large labeled datasets containing labels for all desired entity types together. This paper presents a method for training a single CRF extractor from multiple datasets with disjoint or partially overlapping sets of entity types. Our approach employs marginal likelihood training to insist on labels that are present in the data, while filling in “missing labels”. This allows us to leverage all the available data within a single model. In experimental results on the Biocreative V CDR (chemicals/diseases), Biocreative VI ChemProt (chemicals/proteins) and MedMentions (19 entity types) datasets, we show that joint training on multiple datasets improves NER F1 over training in isolation, and our methods achieve state-of-the-art results.

We present a corpus that encompasses the complete history of conversations between contributors to Wikipedia, one of the largest online collaborative communities. By recording the intermediate states of conversations - including not only comments and replies, but also their modifications, deletions and restorations - this data offers an unprecedented view of online conversation. Our framework is designed to be language agnostic, and we show that it extracts high quality data in both Chinese and English. This level of detail supports new research questions pertaining to the process (and challenges) of large-scale online collaboration. We illustrate the corpus’ potential with two case studies on English Wikipedia that highlight new perspectives on earlier work. First, we explore how a person’s conversational behavior depends on how they relate to the discussion’s venue. Second, we show that community moderation of toxic behavior happens at a higher rate than previously estimated.

As the incidence of Alzheimer’s Disease (AD) increases, early detection becomes crucial. Unfortunately, datasets for AD assessment are often sparse and incomplete. In this work, we leverage the multiview nature of a small AD dataset, DementiaBank, to learn an embedding that captures different modes of cognitive impairment. We apply generalized canonical correlation analysis (GCCA) to our dataset and demonstrate the added benefit of using multiview embeddings in two downstream tasks: identifying AD and predicting clinical scores. By including multiview embeddings, we obtain an F1 score of 0.82 in the classification task and a mean absolute error of 3.42 in the regression task. Furthermore, we show that multiview embeddings can be obtained from other datasets as well.

With the recent rise of #MeToo, an increasing number of personal stories about sexual harassment and sexual abuse have been shared online. In order to push forward the fight against such harassment and abuse, we present the task of automatically categorizing and analyzing various forms of sexual harassment, based on stories shared on the online forum SafeCity. For the labels of groping, ogling, and commenting, our single-label CNN-RNN model achieves an accuracy of 86.5%, and our multi-label model achieves a Hamming score of 82.5%. Furthermore, we present analysis using LIME, first-derivative saliency heatmaps, activation clustering, and embedding visualization to interpret neural model predictions and demonstrate how this helps extract features that can help automatically fill out incident reports, identify unsafe areas, avoid unsafe practices, and ‘pin the creeps’.

Abusive language detection models tend to have a problem of being biased toward identity words of a certain group of people because of imbalanced training datasets. For example, “You are a good woman” was considered “sexist” when trained on an existing dataset. Such model bias is an obstacle for models to be robust enough for practical use. In this work, we measure them on models trained with different datasets, while analyzing the effect of different pre-trained word embeddings and model architectures. We also experiment with three mitigation methods: (1) debiased word embeddings, (2) gender swap data augmentation, and (3) fine-tuning with a larger corpus. These methods can effectively reduce model bias by 90-98% and can be extended to correct model bias in other scenarios.

In this paper, we provide empirical evidence based on a rigourously studied mathematical model for bi-populated networks, that a glass ceiling within the field of NLP has developed since the mid 2000s.

Task oriented dialog systems typically first parse user utterances to semantic frames comprised of intents and slots. Previous work on task oriented intent and slot-filling work has been restricted to one intent per query and one slot label per token, and thus cannot model complex compositional requests. Alternative semantic parsing systems have represented queries as logical forms, but these are challenging to annotate and parse. We propose a hierarchical annotation scheme for semantic parsing that allows the representation of compositional queries, and can be efficiently and accurately parsed by standard constituency parsing models. We release a dataset of 44k annotated queries (http://fb.me/semanticparsingdialog), and show that parsing models outperform sequence-to-sequence approaches on this dataset.

Dialogue state tracker is the core part of a spoken dialogue system. It estimates the beliefs of possible user’s goals at every dialogue turn. However, for most current approaches, it’s difficult to scale to large dialogue domains. They have one or more of following limitations: (a) Some models don’t work in the situation where slot values in ontology changes dynamically; (b) The number of model parameters is proportional to the number of slots; (c) Some models extract features based on hand-crafted lexicons. To tackle these challenges, we propose StateNet, a universal dialogue state tracker. It is independent of the number of values, shares parameters across all slots, and uses pre-trained word vectors instead of explicit semantic dictionaries. Our experiments on two datasets show that our approach not only overcomes the limitations, but also significantly outperforms the performance of state-of-the-art approaches.

Current dialogue systems fail at being engaging for users, especially when trained end-to-end without relying on proactive reengaging scripted strategies. Zhang et al. (2018) showed that the engagement level of end-to-end dialogue models increases when conditioning them on text personas providing some personalized back-story to the model. However, the dataset used in Zhang et al. (2018) is synthetic and only contains around 1k different personas. In this paper we introduce a new dataset providing 5 million personas and 700 million persona-based dialogues. Our experiments show that, at this scale, training using personas still improves the performance of end-to-end systems. In addition, we show that other tasks benefit from the wide coverage of our dataset by fine-tuning our model on the data from Zhang et al. (2018) and achieving state-of-the-art results.

Sequence-to-sequence neural generation models have achieved promising performance on short text conversation tasks. However, they tend to generate generic/dull responses, leading to unsatisfying dialogue experience. We observe that in the conversation tasks, each query could have multiple responses, which forms a 1-to-n or m-to-n relationship in the view of the total corpus. The objective function used in standard sequence-to-sequence models will be dominated by loss terms with generic patterns. Inspired by this observation, we introduce a statistical re-weighting method that assigns different weights for the multiple responses of the same query, and trains the common neural generation model with the weights. Experimental results on a large Chinese dialogue corpus show that our method improves the acceptance rate of generated responses compared with several baseline models and significantly reduces the number of generated generic responses.

We propose to generalize language models for conversational speech recognition to allow them to operate across utterance boundaries and speaker changes, thereby capturing conversation-level phenomena such as adjacency pairs, lexical entrainment, and topical coherence. The model consists of a long-short-term memory (LSTM) recurrent network that reads the entire word-level history of a conversation, as well as information about turn taking and speaker overlap, in order to predict each next word. The model is applied in a rescoring framework, where the word history prior to the current utterance is approximated with preliminary recognition results. In experiments in the conversational telephone speech domain (Switchboard) we find that such a model gives substantial perplexity reductions over a standard LSTM-LM with utterance scope, as well as improvements in word error rate.

The paper introduces end-to-end neural network models that tokenize Sanskrit by jointly splitting compounds and resolving phonetic merges (Sandhi). Tokenization of Sanskrit depends on local phonetic and distant semantic features that are incorporated using convolutional and recurrent elements. Contrary to most previous systems, our models do not require feature engineering or extern linguistic resources, but operate solely on parallel versions of raw and segmented text. The models discussed in this paper clearly improve over previous approaches to Sanskrit word segmentation. As they are language agnostic, we will demonstrate that they also outperform the state of the art for the related task of German compound splitting.

We propose a novel strategy to encode the syntax parse tree of sentence into a learnable distributed representation. The proposed syntax encoding scheme is provably information-lossless. In specific, an embedding vector is constructed for each word in the sentence, encoding the path in the syntax tree corresponding to the word. The one-to-one correspondence between these “syntax-embedding” vectors and the words (hence their embedding vectors) in the sentence makes it easy to integrate such a representation with all word-level NLP models. We empirically show the benefits of the syntax embeddings on the Authorship Attribution domain, where our approach improves upon the prior art and achieves new performance records on five benchmarking data sets.

In natural language processing, a common task is to compute the probability of a phrase appearing in a document or to calculate the probability of all phrases matching a given pattern. For instance, one computes affix (prefix, suffix, infix, etc.) probabilities of a string or a set of strings with respect to a probability distribution of patterns. The problem of computing infix probabilities of strings when the pattern distribution is given by a probabilistic context-free grammar or by a probabilistic finite automaton is already solved, yet it was open to compute the infix probabilities in an incremental manner. The incremental computation is crucial when a new query is built from a previous query. We tackle this problem and suggest a method that computes infix probabilities incrementally for probabilistic finite automata by representing all the probabilities of matching strings as a series of transition matrix calculations. We show that the proposed approach is theoretically faster than the previous method and, using real world data, demonstrate that our approach has vastly better performance in practice.

There have been several recent attempts to improve the accuracy of grammar induction systems by bounding the recursive complexity of the induction model. Modern depth-bounded grammar inducers have been shown to be more accurate than early unbounded PCFG inducers, but this technique has never been compared against unbounded induction within the same system, in part because most previous depth-bounding models are built around sequence models, the complexity of which grows exponentially with the maximum allowed depth. The present work instead applies depth bounds within a chart-based Bayesian PCFG inducer, where bounding can be switched on and off, and then samples trees with or without bounding. Results show that depth-bounding is indeed significantly effective in limiting the search space of the inducer and thereby increasing accuracy of resulting parsing model, independent of the contribution of modern Bayesian induction techniques. Moreover, parsing results on English, Chinese and German show that this bounded model is able to produce parse trees more accurately than or competitively with state-of-the-art constituency grammar induction models.

We provide a comprehensive analysis of the interactions between pre-trained word embeddings, character models and POS tags in a transition-based dependency parser. While previous studies have shown POS information to be less important in the presence of character models, we show that in fact there are complex interactions between all three techniques. In isolation each produces large improvements over a baseline system using randomly initialised word embeddings only, but combining them quickly leads to diminishing returns. We categorise words by frequency, POS tag and language in order to systematically investigate how each of the techniques affects parsing quality. For many word categories, applying any two of the three techniques is almost as good as the full combined system. Character models tend to be more important for low-frequency open-class words, especially in morphologically rich languages, while POS tags can help disambiguate high-frequency function words. We also show that large character embedding sizes help even for languages with small character sets, especially in morphologically rich languages.

Web queries with question intent manifest a complex syntactic structure and the processing of this structure is important for their interpretation. Pinter et al. (2016) has formalized the grammar of these queries and proposed semi-supervised algorithms for the adaptation of parsers originally designed to parse according to the standard dependency grammar, so that they can account for the unique forest grammar of queries. However, their algorithms rely on resources typically not available outside of big web corporates. We propose a new BiLSTM query parser that: (1) Explicitly accounts for the unique grammar of web queries; and (2) Utilizes named entity (NE) information from a BiLSTM NE tagger, that can be jointly trained with the parser. In order to train our model we annotate the query treebank of Pinter et al. (2016) with NEs. When trained on 2500 annotated queries our parser achieves UAS of 83.5% and segmentation F1-score of 84.5, substantially outperforming existing state-of-the-art parsers.