We propose a novel search-based approach for greedy coreference resolution, where the mentions are processed in order and added to previous coreference clusters. Our method is distinguished by the use of two functions to make each coreference decision: a pruning function that prunes bad coreference decisions from further consideration, and a scoring function that then selects the best among the remaining decisions. Our framework reduces learning of these functions to rank learning, which helps leverage powerful off-the-shelf rank-learners. We show that our Prune-and-Score approach is superior to using a single scoring function to make both decisions and outperforms several state-of-the-art approaches on multiple benchmark corpora including OntoNotes.

Abstract. In this paper, we address the problem of automatically generating a description of an image from its annotation. Previous approaches either use computer vision techniques to first determine the labels or exploit available descriptions of the training images to either transfer or compose a new description for the test image. However, none of them report results on the effect of incorrect label detection on the quality of the final descriptions generated. With this motivation, we present an approach to generate image descriptions from image annotation and show that with accurate object and attribute detection, human-like descriptions can be generated. Unlike any previous work, we perform an extensive task-based evaluation to analyze our results.

No abstract

Large scale efforts are underway to create dependency treebanks and parsers for Hindi and other Indian languages. Hindi, being a morphologically rich, flexible word order language, brings challenges such as handling non-projectivity in parsing.In this work, we look at non-projectivity in Hyderabad Dependency Treebank (HyDT) for Hindi. Non-projectivity has been analysed from two perspectives: graph properties that restrict non-projectivity and linguistic phenomenon behind non-projectivity in HyDT. Since Hindi has ample instances of non-projectivity (14% of all structures in HyDT are non-projective), it presents a case for an in depth study of this phenomenon for a better insight, from both of these perspectives.We have looked at graph constriants like planarity, gap degree, edge degree and well-nestedness on structures in HyDT. We also analyse non-projectivity in Hindi in terms of various linguistic parameters such as the causes of non-projectivity, its rigidity (possibility of reordering) and whether the reordered construction is the natural one.

Automatically generating meaningful descriptions for images has recently emerged as an important area of research. In this direction, a nearest-neighbour based generative phrase prediction model (PPM) proposed by (Gupta et al. 2012) was shown to achieve state-of-the-art results on PASCAL sentence dataset, thanks to the simultaneous use of three different sources of information (i.e. visual clues, corpus statistics and available descriptions). However, they do not utilize semantic similarities among the phrases that might be helpful in relating semantically similar phrases during phrase relevance prediction. In this paper, we extend their model by considering inter-phrase semantic similarities. To compute similarity between two phrases, we consider similarities among their constituent words determined using WordNet. We also re-formulate their objective function for parameter learning by penalizing each pair of phrases unevenly, in a manner similar to that in structured predictions. Various automatic and human evaluations are performed to demonstrate the advantage of our "semantic phrase prediction model" (SPPM) over PPM.