NAACL 2016 论文列表

We address the problem of automatically finding the parameters of a statistical machine translation system that maximize BLEU scores while ensuring that decoding speed exceeds a minimum value. We propose the use of Bayesian Optimization to efficiently tune the speed-related decoding parameters by easily incorporating speed as a noisy constraint function. The obtained parameter values are guaranteed to satisfy the speed constraint with an associated confidence margin. Across three language pairs and two speed constraint values, we report overall optimization time reduction compared to grid and random search. We also show that Bayesian Optimization can decouple speed and BLEU measurements, resulting in a further reduction of overall optimization time as speed is measured over a small subset of sentences.

We propose a novel neural network model for machine reading, DER Network, which explicitly implements a reader building dynamic meaning representations for entities by gathering and accumulating information around the entities as it reads a document. Evaluated on a recent large scale dataset (Hermann et al., 2015), our model exhibits better results than previous research, and we find that max-pooling is suited for modeling the accumulation of information on entities. Further analysis suggests that our model can put together multiple pieces of information encoded in different sentences to answer complicated questions. Our code for the model is available at https://github. com/soskek/der-network

Representation and learning of commonsense knowledge is one of the foundational problems in the quest to enable deep language understanding. This issue is particularly challenging for understanding casual and correlational relationships between events. While this topic has received a lot of interest in the NLP community, research has been hindered by the lack of a proper evaluation framework. This paper attempts to address this problem with a new framework for evaluating story understanding and script learning: the 'Story Cloze Test'. This test requires a system to choose the correct ending to a four-sentence story. We created a new corpus of ~50k five-sentence commonsense stories, ROCStories, to enable this evaluation. This corpus is unique in two ways: (1) it captures a rich set of causal and temporal commonsense relations between daily events, and (2) it is a high quality collection of everyday life stories that can also be used for story generation. Experimental evaluation shows that a host of baselines and state-of-the-art models based on shallow language understanding struggle to achieve a high score on the Story Cloze Test. We discuss these implications for script and story learning, and offer suggestions for deeper language understanding.

We address relation classification in the context of slot filling, the task of finding and evaluating fillers like “Steve Jobs” for the slot X in “X founded Apple”. We propose a convolutional neural network which splits the input sentence into three parts according to the relation arguments and compare it to state-ofthe-art and traditional approaches of relation classification. Finally, we combine different methods and show that the combination is better than individual approaches. We also analyze the effect of genre differences on performance.

,

Access to word-sentiment associations is useful for many applications, including sentiment analysis, stance detection, and linguistic analysis. However, manually assigning fine-grained sentiment association scores to words has many challenges with respect to keeping annotations consistent. We apply the annotation technique of Best-Worst Scaling to obtain real-valued sentiment association scores for words and phrases in three different domains: general English, English Twitter, and Arabic Twitter. We show that on all three domains the ranking of words by sentiment remains remarkably consistent even when the annotation process is repeated with a different set of annotators. We also, for the first time, determine the minimum difference in sentiment association that is perceptible to native speakers of a language.

We present an approach to the new task of opinion holder and target extraction on opinion compounds. Opinion compounds (e.g. user rating or victim support) are noun compounds whose head is an opinion noun. We do not only examine features known to be effective for noun compound analysis, such as paraphrases and semantic classes of heads and modifiers, but also propose novel features tailored to this new task. Among them, we examine paraphrases that jointly consider holders and targets, a verb detour in which noun heads are replaced by related verbs, a global head constraint allowing inferencing between different compounds, and the categorization of the sentiment view that the head conveys.

This paper presents a clustering approach that simultaneously identifies product features and groups them into aspect categories from online reviews. Unlike prior approaches that first extract features and then group them into categories, the proposed approach combines feature and aspect discovery instead of chaining them. In addition, prior work on feature extraction tends to require seed terms and focus on identifying explicit features, while the proposed approach extracts both explicit and implicit features, and does not require seed terms. We evaluate this approach on reviews from three domains. The results show that it outperforms several state-of-the-art methods on both tasks across all three domains.

We examine different features and classifiers for the categorization of opinion words into actor and speaker view. To our knowledge, this is the first comprehensive work to address sentiment views on the word level taking into consideration opinion verbs, nouns and adjectives. We consider many high-level features requiring only few labeled training data. A detailed feature analysis produces linguistic insights into the nature of sentiment views. We also examine how far global constraints between different opinion words help to increase classification performance. Finally, we show that our (prior) word-level annotation correlates with contextual sentiment views.

Embeddings are generic representations that are useful for many NLP tasks. In this paper, we introduce DENSIFIER, a method that learns an orthogonal transformation of the embedding space that focuses the information relevant for a task in an ultradense subspace of a dimensionality that is smaller by a factor of 100 than the original space. We show that ultradense embeddings generated by DENSIFIER reach state of the art on a lexicon creation task in which words are annotated with three types of lexical information - sentiment, concreteness and frequency. On the SemEval2015 10B sentiment analysis task we show that no information is lost when the ultradense subspace is used, but training is an order of magnitude more efficient due to the compactness of the ultradense space.

In this paper, we examine the impact of employing contextual, structural information from a tree-structured document set to derive a language model. Our results show that this information significantly improves the accuracy of the resultant model.

We propose a framework for devising empirically testable algorithms for bridging the communication gap between humans and robots. We instantiate our framework in the context of a problem setting in which humans give instructions to robots using unrestricted natu-ral language commands, with instruction sequences being subservient to building complex goal configurations in a blocks world. We show how one can collect meaningful training data and we propose three neural architectures for interpreting contextually grounded natural language commands. The proposed architectures allow us to correctly understand/ground the blocks that the robot should move when instructed by a human who uses unrestricted language. The architectures have more diffi-culty in correctly understanding/grounding the spatial relations required to place blocks correctly, especially when the blocks are not easily identifiable.

We present a corpus of 2,380 natural language queries paired with machine readable formulae that can be executed against world wide geographic data of the OpenStreetMap (OSM) database. We use the corpus to learn an accurate semantic parser that builds the basis of a natural language interface to OSM. Furthermore, we use response-based learning on parser feedback to adapt a statistical machine translation system for multilingual database access to OSM. Our framework allows to map fuzzy natural language expressions such as “nearby”, “north of”, or “in walking distance” to spatial polygons on an interactive map. Furthermore, it combines syntactic complexity and compositionality with a reasonable lexical variability of queries, making it an interesting new publicly available dataset for research on semantic parsing.

Language generation from purely semantic representations is a challenging task. This paper addresses generating English from the Ab-stract Meaning Representation (AMR), consisting of re-entrant graphs whose nodes are concepts and edges are relations. The new method is trained statistically from AMR-annotated English and consists of two major steps: (i) generating an appropriate spanning tree for the AMR, and (ii) applying tree-to-string transducers to generate English. The method relies on discriminative learning and an argument realization model to overcome data sparsity. Initial tests on held-out data show good promise despite the complexity of the task. The system is available open-source as part of JAMR at:

We propose an end-to-end, domain-independent neural encoder-aligner-decoder model for selective generation, i.e., the joint task of content selection and surface realization. Our model first encodes a full set of over-determined database event records via an LSTM-based recurrent neural network, then utilizes a novel coarse-to-fine aligner to identify the small subset of salient records to talk about, and finally employs a decoder to generate free-form descriptions of the aligned, selected records. Our model achieves the best selection and generation results reported to-date (with 59% relative improvement in generation) on the benchmark WeatherGov dataset, despite using no specialized features or linguistic resources. Using an improved k-nearest neighbor beam filter helps further. We also perform a series of ablations and visualizations to elucidate the contributions of our key model components. Lastly, we evaluate the generalizability of our model on the RoboCup dataset, and get results that are competitive with or better than the state-of-the-art, despite being severely data-starved.

This paper introduces a new annotated corpus based on an existing informal text corpus: the NUS SMS Corpus (Chen and Kan, 2013). The new corpus includes 76,490 noun phrases from 26,500 SMS messages, annotated by university students. We then explored several graphical models, including a novel variant of the semi-Markov conditional random fields (semi-CRF) for the task of noun phrase chunking. We demonstrated through empirical evaluations on the new dataset that the new variant yielded similar accuracy but ran in significantly lower running time compared to the conventional semi-CRF.

This paper addresses the problem of comment classification in community Question Answering. Following the state of the art, we approach the task with a global inference process to exploit the information of all comments in the answer-thread in the form of a fully connected graph. Our contribution comprises two novel joint learning models that are on-line and integrate inference within learning. The first one jointly learns two node - and edge -level MaxEnt classifiers with stochastic gradient descent and integrates the inference step with loopy belief propagation. The second model is an instance of fully connected pairwise CRFs (FCCRF). The FCCRF model significantly outperforms all other approaches and yields the best results on the task to date. Crucial elements for its success are the global normalization and an Ising-like edge potential.

In many languages, sparse availability of resources causes numerous challenges for textual analysis tasks. Text classification is one of such standard tasks that is hindered due to limited availability of label information in lowresource languages. Transferring knowledge (i.e. label information) from high-resource to low-resource languages might improve text classification as compared to the other approaches like machine translation. We introduce BRAVE (Bilingual paRAgraph VEctors), a model to learn bilingual distributed representations (i.e. embeddings) of words without word alignments either from sentencealigned parallel or label-aligned non-parallel document corpora to support cross-language text classification. Empirical analysis shows that classification models trained with our bilingual embeddings outperforms other stateof-the-art systems on three different crosslanguage text classification tasks.

While neural networks have been successfully applied to many NLP tasks the resulting vector-based models are very difficult to interpret. For example it's not clear how they achieve {\em compositionality}, building sentence meaning from the meanings of words and phrases. In this paper we describe four strategies for visualizing compositionality in neural models for NLP, inspired by similar work in computer vision. We first plot unit values to visualize compositionality of negation, intensification, and concessive clauses, allow us to see well-known markedness asymmetries in negation. We then introduce three simple and straightforward methods for visualizing a unit's {\em salience}, the amount it contributes to the final composed meaning: (1) gradient back-propagation, (2) the variance of a token from the average word node, (3) LSTM-style gates that measure information flow. We test our methods on sentiment using simple recurrent nets and LSTMs. Our general-purpose methods may have wide applications for understanding compositionality and other semantic properties of deep networks , and also shed light on why LSTMs outperform simple recurrent nets,