Eduard Hovy

Carnegie Mellon University | CMU · Language Technologies Institute

Text style transfer aims to controllably generate text with targeted stylistic changes while maintaining core meaning from the source sentence constant. Many of the existing style transfer benchmarks primarily focus on individual high-level semantic changes (e.g. positive to negative), which enable controllability at a high level but do not offer f...

In this work, we propose a new generative model that is capable of automatically decoupling global and local representations of images in an entirely unsupervised setting. The proposed model utilizes the variational auto-encoding framework to learn a (low-dimensional) vector of latent variables to capture the global information of an image, which i...

The LoReHLT16 evaluation challenged participants to extract Situation Frames (SFs)—structured descriptions of humanitarian need situations—from monolingual Uyghur text. The ARIEL-CMU SF detector combines two classification paradigms, a manually curated keyword-spotting system and a machine learning classifier. These were applied by translating the...

In this paper, we propose a probabilistic parsing model, which defines a proper conditional probability distribution over non-projective dependency trees for a given sentence, using neural representations as inputs. The neural network architecture is based on bi-directional LSTM-CNNs which benefits from both word- and character-level representation...

Cloze test is widely adopted in language exams to evaluate students' language proficiency. In this paper, we propose the first large-scale human-designed cloze test dataset CLOTH, in which the questions were used in middle-school and high-school language exams. With the missing blanks carefully created by teachers and candidate choices purposely de...

Entity-based ranking systems often employ entity linking systems to align entities to query and documents. Previously, entity linking systems were not designed specifically for search engines and were mostly used as a preprocessing step. This work presents JointSem, a joint semantic ranking system that combines query entity linking and entity-based...

Explaining underlying causes or effects about events is a challenging but valuable task. We define a novel problem of generating explanations of a time series event by (1) searching cause and effect relationships of the time series with textual data and (2) constructing a connecting chain between them to generate an explanation. To detect causal fe...

Portmanteaus are a word formation phenomenon where two words are combined to form a new word. We propose character-level neural sequence-to-sequence (S2S) methods for the task of portmanteau generation that are end-to-end-trainable, language independent, and do not explicitly use additional phonetic information. We propose a noisy-channel-style mod...

Variations in writing styles are commonly used to adapt the content to a specific context, audience, or purpose. However, applying stylistic variations is still by and large a manual process, and there have been little efforts towards automating it. In this paper we explore automated methods to transform text from modern English to Shakespearean En...

Learning meaningful representations that maintain the content necessary for a particular task while filtering away detrimental variations is a problem of great interest in machine learning. In this paper, we tackle the problem of learning representations invariant to a specific factor or trait of data, leading to better generalization. The represen...

Reward augmented maximum likelihood (RAML) is a simple and effective learning framework to directly optimize towards the reward function in structured prediction tasks. RAML incorporates task-specific reward by performing maximum-likelihood updates on candidate outputs sampled according to an exponentiated payoff distribution, which gives higher pr...

Type-level word embeddings use the same set of parameters to represent all instances of a word regardless of its context, ignoring the inherent lexical ambiguity in language. Instead, we embed semantic concepts (or synsets) as defined in WordNet and represent a word token in a particular context by estimating a distribution over relevant semantic c...

Knowledge bases are important resources for a variety of natural language processing tasks but suffer from incompleteness. We propose a novel embedding model, \emph{ITransF}, to perform knowledge base completion. Equipped with a sparse attention mechanism, ITransF discovers hidden concepts of relations and transfer statistical strength through the...

We present RACE, a new dataset for benchmark evaluation of methods in the reading comprehension task. Collected from the English exams for middle and high school Chinese students in the age range between 12 to 18, RACE consists of 28,000+ passages and near 100,000 questions generated by human experts (English instructors), and covers a variety of t...

The detection of expressions of sentiment in online text has become a popular Natural Language Processing application. The task is commonly defined as identifying the words or phrases in a given fragment of text in which the reader understands that the author expresses some person’s positive, negative, or perhaps neutral attitude toward a topic. Th...

In this paper, we propose to equip Generative Adversarial Networks with the ability to produce direct energy estimates for samples.Specifically, we propose a flexible adversarial training framework, and prove this framework not only ensures the generator converges to the true data distribution, but also enables the discriminator to retain the densi...

Curriculum Learning emphasizes the order of training instances in a computational learning setup. The core hypothesis is that simpler instances should be learned early as building blocks to learn more complex ones. Despite its usefulness, it is still unknown how exactly the internal representation of models are affected by curriculum learning. In t...

Dropout, a simple and effective way to train deep neural networks, has led to a number of impressive empirical successes and spawned many recent theoretical investigations. However, the gap between dropout's training and inference phases, introduced due to tractability considerations, has largely remained under-appreciated. In this work, we first f...

Coreference resolution is one of the first stages in deep language understanding and its importance has been well recognized in the natural language processing community. In this paper, we propose a generative, unsupervised ranking model for entity coreference resolution by introducing resolution mode variables. Our unsupervised system achieves 58....

Hoax calls annually cost law enforcement and security agencies over a billion dollars, and sometimes lives. Bogus bomb threats, “swatting” calls to the police, hoax calls to the coast guard etc. cause these agencies to respond, deploying personnel and resources needlessly. The response itself could cause direct danger to innocent citizens, while al...

Combining deep neural networks with structured logic rules is desirable to harness flexibility and reduce unpredictability of the neural models. We propose a general framework capable of enhancing various types of neural networks (e.g., CNNs and RNNs) with declarative first-order logic rules. Specifically, we develop an iterative distillation metho...

State-of-the-art sequence labeling systems traditionally require large amounts of task-specific knowledge in the form of hand-crafted features and data pre-processing. In this paper, we introduce a novel neutral network architecture that benefits from both word- and character-level representations automatically, by using combination of bidirectiona...

We describe two new related resources that facilitate modelling of general knowledge reasoning in 4th grade science exams. The first is a collection of curated facts in the form of tables, and the second is a large set of crowd-sourced multiple-choice questions covering the facts in the tables. Through the setup of the crowd-sourced annotation task...

This report documents the program and the outcomes of Dagstuhl Seminar 15512 "Debating Technologies". The seminar brought together leading researchers from computational linguistics, information retrieval, semantic web, and database communities to discuss the possibilities, implications, and necessary actions for the establishment of a new interdis...

Many NLP systems use dependency parsers as critical components. Jonit learning parsers usually achieve better parsing accuracies than two-stage methods. However , classical joint parsing algorithms significantly increase computational complexity , which makes joint learning impractical. In this paper, we proposed an efficient dependency parsing alg...

This report documents the program and the outcomes of Dagstuhl Perspectives Workshop 11331 "The Future of Research Communication". The purpose of the workshop was to bring together researchers from these different disciplines, whose core research goal is changing the formats, standards, and means by which we communicate science.

In this paper, we described possible directions for deeper understanding, helping bridge the gap between psychology / cognitive science and computational approaches in sentiment/opinion analysis literature. We focus on the opinion holder's underlying needs and their resultant goals, which, in a utilitarian model of sentiment, provides the basis for...

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...

It is commonly accepted that machine translation is a more complex task than part of speech tagging. But how much more complex? In this paper we make an attempt to develop a general framework and methodology for computing the informational and/or processing complexity of NLP applications and tasks. We define a universal framework akin to a Turning...

Recursive neural models, which use syntactic parse trees to recursively generate representations bottom-up from parse children, are a popular new architecture, promising to capture structural properties like the scope of negation or long-distance semantic dependencies. But understanding exactly which tasks this parse-based method is appropriate for...

In Natural Language Processing or Computational Linguistics (NLP or CL), researchers assume almost universally that speakers hold some affective value or sentiment with regard to (some aspects of) a topic such as a film or camera, that this sentiment has a fixed value (typically, something like good or bad), and that the sentiment is expressed in t...

Event Mention detection is the first step in textual event understanding. Proper evaluation is important for modern natural language processing tasks. In this paper, we present our evaluation algorithm and results during the Event Mention Evaluation pilot study. We analyze the problems of evaluating multiple event mention attributes and discontinuo...

In this paper, we propose a novel approach for Word Sense Disambiguation (WSD) of verbs that can be applied directly in the event mention detection task to classify event types. By using the PropStore, a database of relations between words, our approach disambiguates senses of verbs by utilizing the information of verbs that appear in similar synta...

We have proposed a simple, effective and fast method named retrofitting to improve word vectors using word relation knowledge found in semantic lexicons constructed either automatically or by humans. Retrofitting is used as a post-processing step to improve vector quality and is simpler to use than other approaches that use semantic information whi...

While it has long been believed in psychology that weather somehow influences human's mood, the debates have been going on for decades about how they are correlated. In this paper, we try to study this long-lasting topic by harnessing a new source of data compared from traditional psychological researches: Twitter. We analyze 2 years' twitter data...

While it has long been believed in psychology that weather somehow influences human's mood, the debates have been going on for decades about how they are correlated. In this paper, we try to study this long-lasting topic by harnessing a new source of data compared from traditional psychological researches: Twitter. We analyze 2 years' twitter data...

We propose a new post-editing method for statistical machine translation. The method acquires translation rules automatically as translation knowledge from a parallel corpus without depending on linguistic tools. The translation rules, which are acquired based on Intuitive Common Parts Continuum (ICPC), can deal with the correspondence of the globa...

Most computational approaches to metaphor detection try to leverage either conceptual metaphor mappings or selectional preferences. Both require extensive knowledge of the mappings/preferences in question, as well as sufficient data for all involved conceptual domains. Creating these resources is expensive and often limits the scope of these system...

Smart government is possible only if the security of sensitive data can be assured. The more knowledgeable government officials and citizens are about cybersecurity, the better are the chances that government data is not compromised or abused. In this paper, we present two systems under development that aim at improving cybersecurity education. Fir...

As described in this paper, we propose a new automatic evaluation metric for machine translation. Our metric is based on chunking between the reference and candidate translation. Moreover, we apply a prize based on sentence-length to the metric, dissimilar from penalties in BLEU or NIST. We designate this metric as Automatic Evaluation of Machine T...

Traditional models of distributional semantics suffer from computational issues such as data sparsity for individual lexemes and complexities of modeling semantic composition when dealing with structures larger than single lexical items. In this work, we present a frequencydriven paradigm for robust distributional semantics in terms of semantically...

BLANC is a link-based coreference evaluation metric for measuring the quality of coreference systems on gold mentions. This paper extends the original BLANC ("BLANC-gold" henceforth) to system mentions, removing the gold mention assumption. The proposed BLANC falls back seamlessly to the original one if system mentions are identical to gold mention...

Consumers' purchase decisions are increasingly influenced by user-generated online reviews. Accordingly, there has been growing concern about the potential for posting deceptive opinion spam - fictitious reviews that have been deliberately written to sound authentic, to deceive the reader. In this paper, we explore generalized approaches for identi...

While user attribute extraction on social media has received considerable attention, existing approaches, mostly supervised, encounter great difficulty in obtaining gold standard data and are therefore limited to predicting unary predicates (e.g., gender). In this paper, we present a weaklysupervised approach to user profile extraction from Twitter...

The definitions of two coreference scoring metrics-B3 and CEAF-are underspecified with respect to predicted, as opposed to key (or gold) mentions. Several variations have been proposed that manipulate either, or both, the key and predicted mentions in order to get a one-to-one mapping. On the other hand, the metric BLANC was, until recently, limite...

In this paper, we address the problem of discovering topically meaningful, yet compact (densely connected) communities in a social network. Assuming the social network to be an integer-weighted graph (where the weights can be intuitively defined as the number of common friends, followers, documents exchanged, etc.), we transform the social network...

This paper proposes an evaluation scheme to measure the performance of a system that detects hierarchical event structure for event coreference resolution. We show that each system output is represented as a forest of unordered trees, and introduce the notion of conceptual event hierarchy to simplify the evaluation process. We enumerate the desider...

This paper describes the methodology for testing the performance of Machine Reading systems through Question Answering and Reading Comprehension Tests. This was the attempt of the QA4MRE challenge which was run as a Lab at CLEF 2011–2013. The traditional QA task was replaced by a new Machine Reading task, whose intention was to ask questions that r...

It has long been a dream to have available a single, centralized, semantic thesaurus or terminology taxonomy to support research in a variety of fields. Much human and computational effort has gone into constructing such resources, including the original WordNet and subsequent wordnets in various languages. To produce such resources one has to over...

Paraphrases are sentences or phrases that convey the same meaning using different wording. Although the logical definition of paraphrases requires strict semantic equivalence, linguistics accepts a broader, approximate, equivalence-thereby allowing far more examples of "quasiparaphrase." But approximate equivalence is hard to define. Thus, the phen...

This paper uses a crowd-sourced definition of a speech phenomenon we have called "focus". Given sentences, text and speech, in isolation and in context, we asked annotators to identify what we term the "focus" word. We present their consistency in identifying the focused word, when presented with text or speech stimuli. We then build models to show...

The English's possessive construction occurs frequently in text and can encode several different semantic relations; however, it has received limited attention from the computational linguistics community. This paper describes the creation of a semantic relation inventory covering the use of 's, an inter-annotator agreement study to calculate how w...

In this paper we present a novel approach to modelling distributional semantics that represents meaning as distributions over relations in syntactic neighborhoods. We argue that our model approximates meaning in compositional configurations more effectively than standard distributional vectors or bag-of-words models. We test our hypothesis on the p...

Participants in online decision making environments assume different roles. Especially in contentious discussions, the outcome often depends critically on the discussion leader(s). Recent work on automated leadership analysis has focused on collaborations where all the participants have the same goal. In this paper we focus on contentious discussio...

Random walks is one of the most popular ideas in computer science. A critical assumption in random walks is that the probability of the walk being at a given vertex at a time instance converges to a limit independent of the start state. While this makes it computationally efficient to solve, it limits their use to incorporate label information. In...

Recent years have seen a great deal of work that exploits collaborative, semi-structured content for Artificial Intelligence (AI) and Natural Language Processing (NLP). This special issue of the Artificial Intelligence Journal presents a variety of state-of-the-art contributions, each of which illustrates the substantial impact that work on leverag...

In this paper, we propose a walk-based graph kernel that generalizes the notion of treekernels to continuous spaces. Our proposed approach subsumes a general framework for word-similarity, and in particular, provides a flexible way to incorporate distributed representations. Using vector representations, such an approach captures both distributiona...

We propose new automatic evaluation metric to evaluate machine translation. Different from most similar metrics, our proposed metric does not depend heavily on sentence length. In most metrics based on f-measure comparisons of reference and candidate translations, the relative weight of each mismatched word in short sentences is larger than it in l...

Recent years have seen a great deal of work that exploits collaborative, semi-structured content for Artificial Intelligence (AI) and Natural Language Processing (NLP). This special issue of the Artificial Intelligence Journal presents a variety of state-of-the-art contributions, each of which illustrates the substantial impact that work on leverag...

For many NLP tasks, EM-trained HMMs are the common models. However, in order to escape local maxima and find the best model, we need to start with a good initial model. Researchers suggested repeated random restarts or constraints that guide the model evolution. Neither approach is ideal. Restarts are time-intensive, and most constraint-based appro...

Sex trafficking is the process and means of using force, fraud, or coercion to obtain and compel men, women and children into commercial sexual exploitation. Prevalent in both international and domestic spheres, this form of human trafficking constitutes a serious crime. Traffickers use a variety of means to advertise the illicit sexual services of...