We present a scalable joint language model designed to utilize fine-grain syntactic tags. We discuss challenges such a design faces and describe our solutions that scale well to large tagsets and corpora. We advocate the use of relatively simple tags that do not require deep linguistic knowledge of the language but provide more structural information than POS tags and can be derived from automatically generated parse trees – a combination of properties that allows easy adoption of this model for new languages. We propose two fine-grain tagsets and evaluate our model using these tags, as well as POS tags and SuperARV tags in a speech recognition task and discuss future directions. 1

It is well known that parsing accuracies drop significantly on out-of-domain data. What is less known is that some parsers suffer more from domain shifts than others. We show that dependency parsers have more difficulty parsing questions than constituency parsers. In particular, deterministic shift-reduce dependency parsers, which are of highest interest for practical applications because of their linear running time, drop to 60 % labeled accuracy on a question test set. We propose an uptraining procedure in which a deterministic parser is trained on the output of a more accurate, but slower, latent variable constituency parser (converted to dependencies). Uptraining with 100K unlabeled questions achieves results comparable to having 2K labeled questions for training. With 100K unlabeled and 2K labeled questions, uptraining is able to improve parsing accuracy to 84%, closing the gap between in-domain and out-of-domain performance. 1

We show that the automatically induced latent variable grammars of Petrov et al. (2006) vary widely in their underlying representations, depending on their EM initialization point. We use this to our advantage, combining multiple automatically learned grammars into an unweighted product model, which gives significantly improved performance over state-ofthe-art individual grammars. In our model, the probability of a constituent is estimated as a product of posteriors obtained from multiple grammars that differ only in the random seed used for initialization, without any learning or tuning of combination weights. Despite its simplicity, a product of eight automatically learned grammars improves parsing accuracy from 90.2 % to 91.8 % on English, and from 80.3 % to 84.5 % on German. 1

Text simplification is the process of changing vocabulary and grammatical structure to create a more accessible version of the text while maintaining the underlying information and content. Automated tools for text simplification are a practical way to make large corpora of text accessible to a wider audience lacking high levels of fluency in the corpus language. In this work, we investigate the potential of Simple Wikipedia to assist automatic text simplification by building a statistical classification system that discriminates simple English from ordinary English. Most text simplification systems are based on hand-written rules (e.g., PEST (Carroll et al., 1999) and its module SYSTAR (Canning et al., 2000)), and therefore face limitations scaling and transferring across domains. The potential for using Simple Wikipedia for text simplification is significant; it contains nearly 60,000 articles with revision histories and aligned articles to ordinary English Wikipedia. Using articles from Simple Wikipedia and ordinary Wikipedia, we evaluated different classifiers and feature sets to identify the most discriminative features of simple English for use across domains. These findings help further understanding of what makes text simple and can be applied as a tool to help writers craft simple text. 1

This paper presents a study of the impact of using simple and complex morphological clues to improve the classification of rare and unknown words for parsing. We compare this approach to a language-independent technique often used in parsers which is based solely on word frequencies. This study is applied to three languages that exhibit different levels of morphological expressiveness: Arabic, French and English. We integrate information about Arabic affixes and morphotactics into a PCFG-LA parser and obtain stateof-the-art accuracy. We also show that these morphological clues can be learnt automatically from an annotated corpus. 1

Out-of-vocabulary (OOV) words represent an important source of error in large vocabulary continuous speech recognition (LVCSR) systems. These words cause recognition failures, which propagate through pipeline systems impacting the performance of downstream applications. The detection of OOV regions in the output of a LVCSR system is typically addressed as a binary classification task, where each region is independently classified using local information. In this paper, we show that jointly predicting OOV regions, and including contextual information from each region, leads to substantial improvement in OOV detection. Compared to the state-of-the-art, we reduce the missed OOV rate from 42.6 % to 28.4 % at 10 % false alarm rate. 1

In this paper, we present a novel approach to enhance hierarchical phrase-based machine translation systems with linguistically motivated syntactic features. Rather than directly using treebank categories as in previous studies, we learn a set of linguistically-guided latent syntactic categories automatically from a source-side parsed, word-aligned parallel corpus, based on the hierarchical structure among phrase pairs as well as the syntactic structure of the source side. In our model, each X nonterminal in a SCFG rule is decorated with a real-valued feature vector computed based on its distribution of latent syntactic categories. These feature vectors are utilized at decoding time to measure the similarity between the syntactic analysis of the source side and the syntax of the SCFG rules that are applied to derive translations. Our approach maintains the advantages of hierarchical phrase-based translation systems while at the same time naturally incorporates soft syntactic constraints.

In this paper, we offer broad insight into the underperformance of Arabic constituency parsing by analyzing the interplay of linguistic phenomena, annotation choices, and model design. First, we identify sources of syntactic ambiguity understudied in the existing parsing literature. Second, we show that although the Penn Arabic Treebank is similar to other treebanks in gross statistical terms, annotation consistency remains problematic. Third, we develop a human interpretable grammar that is competitive with a latent variable PCFG. Fourth, we show how to build better models for three different parsers. Finally, we show that in application settings, the absence of gold segmentation lowers parsing performance by 2–5 % F1. 1

We study self-training with products of latent variable grammars in this paper. We show that increasing the quality of the automatically parsed data used for self-training gives higher accuracy self-trained grammars. Our generative self-trained grammars reach F scores of 91.6 on the WSJ test set and surpass even discriminative reranking systems without selftraining. Additionally, we show that multiple self-trained grammars can be combined in a product model to achieve even higher accuracy. The product model is most effective when the individual underlying grammars are most diverse. Combining multiple grammars that were self-trained on disjoint sets of unlabeled data results in a final test accuracy of 92.5 % on the WSJ test set and 89.6 % on our Broadcast News test set. 1

In the face of sparsity, statistical models are often interpolated with lower order (backoff) models, particularly in Language Modeling. In this paper, we argue that there is a relation between the higher order and the backoff model that must be satisfied in order for the interpolation to be effective. We show that in n-gram models, the relation is trivially held, but in models that allow arbitrary clustering of context (such as decision tree models), this relation is generally not satisfied. Based on this insight, we also propose a generalization of linear interpolation which significantly improves the performance of a decision tree language model. Note the context space for this function, w i−1 1 is arbitrarily long, necessitating some independence assumption, which usually consists of reducing the relevant context to n − 1 immediately preceding tokens: p(wi|w i−1 1) ≈ p(wi|w i−1 i−n+1) These distributions are typically estimated from observed counts of n-grams wi i−n+1 in the training data. The context space is still far too large; therefore, the models are recursively smoothed using lower order distributions. For instance, in a widely used n-gram LM, the probabilities are estimated as follows: 1