Articles

For phylogenetic inference, linguistic typol- ogy is a promising alternative to lexical evi- dence because it allows us to compare an ar- bitrary pair of languages. A challenging prob- lem with typology-based phylogenetic infer- ence is that the changes of typological fea- tures over time are less intuitive than those …...

One of the weaknesses of current supervised word sense disambiguation (WSD) systems is that they only treat a word as a discrete en- tity. However, a continuous-space represen- tation of words (word embeddings) can pro- vide valuable information and thus improve generalization accuracy. Since word embed- dings are typically obtained …...

This paper describes a joint model of word segmentation and phonological alternations, which takes unsegmented utterances as input and infers word segmentations and underlying phonological representations. The model is a Maximum Entropy or log-linear model, which can express a probabilistic version of Opti- mality Theory (OT; Prince and Smolensky (2004)), …...

We propose a method for simultaneously translating from a single source language to multiple target languages T1, T2, etc. The mo- tivation behind this method is that if we only have a weak language model for T1 and trans- lations in T1 and T2 are associated, we can use the …...

Marketing materials such as flyers and other infographics are a vast online resource. In a number of industries, such as the com- mercial real estate industry, they are in fact the only authoritative source of information. Companies attempting to organize commer- cial real estate inventories spend a significant amount of …...

A large part of human communication in- volves referring to entities in the world and often these entities are objects that are visu- ally present for the interlocutors. A system that aims to resolve such references needs to tackle a complex task: objects and their vi- sual features need to …...

This paper presents a novel technique for empty category (EC) detection using dis- tributed word representations. A joint model is learned from the labeled data to map both the distributed representations of the contexts of ECs and EC types to a low dimensional space. In the testing phase, the context …...

We present a discriminative model for de- tecting disfluencies in spoken language tran- scripts. Structurally, our model is a semi- Markov conditional random field with features targeting characteristics unique to speech re- pairs. This gives a significant performance improvement over standard chain-structured CRFs that have been employed in past work. …...

Word embeddings have been found useful for many NLP tasks, including part-of-speech tagging, named entity recognition, and pars- ing. Adding multilingual context when learn- ing embeddings can improve their quality, for example via canonical correlation analysis (CCA) on embeddings from two languages. In this paper, we extend this idea to …...

Several techniques have recently been pro- posed for training “self-normalized” discrimi- native models. These attempt to find parameter settings for which unnormalized model scores approximate the true label probability. How- ever, the theoretical properties of such tech- niques (and of self-normalization generally) have not been investigated. This paper exam- ines …...

The task of Named Entity Disambiguation is to map entity mentions in the document to their correct entries in some knowledge base. We present a novel graph-based dis- ambiguation approach based on Personalized PageRank (PPR) that combines local and global evidence for disambiguation and ef- fectively filters out noise introduced …...

Monolingual alignment models have been shown to boost the performance of question answering systems by ”bridging the lexical chasm” between questions and answers. The main limitation of these approaches is that they require semistructured training data in the form of question-answer pairs, which is diffi- cult to obtain in specialized …...

Research on entity linking has considered a broad range of text, including newswire, blogs and web documents in multiple languages. However, the problem of entity linking for spoken language remains unexplored. Spo- ken language obtained from automatic speech recognition systems poses different types of challenges for entity linking; transcription er- …...

This work improves monolingual sentence alignment for text simplification, specifically for text in standard and simple Wikipedia. We introduce a method that improves over past efforts by using a greedy (vs. ordered) search over the document and a word-level se- mantic similarity score based on Wiktionary (vs. WordNet) that also …...

We present an intuitive and effective method for inducing style scores on words and phrases. We exploit signal in a phrase’s rate of occurrence across stylistically contrasting cor- pora, making our method simple to implement and efficient to scale. We show strong results both intrinsically, by correlation with human judgements, …...

A portmanteau is a type of compound word that fuses the sounds and meanings of two component words; for example, “frenemy” (friend + enemy) or “smog” (smoke + fog). We develop a system, including a novel mul- titape FST, that takes an input of two words and outputs possible portmanteaux. …...

We present a novel response generation sys- tem that can be trained end to end on large quantities of unstructured Twitter conversa- tions. A neural network architecture is used to address sparsity issues that arise when in- tegrating contextual information into classic statistical models, allowing the system to take into …...

Inferring latent attributes of online users has many applications in public health, politics, and marketing. Most existing approaches rely on supervised learning algorithms, which re- quire manual data annotation and therefore are costly to develop and adapt over time. In this paper, we propose a lightly supervised approach based on …...

We address the problem of automatically aligning natural language sentences with cor- responding video segments without any direct supervision. Most existing algorithms for in- tegrating language with videos rely on hand- aligned parallel data, where each natural lan- guage sentence is manually aligned with its corresponding image or video segment. …...

We extend the SKIP-GRAM model of Mikolov et al. (2013a) by taking visual information into account. Like SKIP-GRAM, our multimodal models (MMSKIP-GRAM) build vector-based word representations by learning to predict linguistic contexts in text corpora. However, for a restricted set of words, the models are also exposed to visual representations …...