Cube Pruning is a fast method to explore the search space of a beam decoder. In this paper we present two modifications of the algorithm that aim to improve the speed and reduce the amount of memory needed at execution time. We show that, in applications where Cube Pruning is applied to a monotonic search space, the proposed algorithms retrieve the same K-best set with lower complexity. When tested on an application where the search space is approximately monotonic (Machine Translation with Language Model features), we show that the proposed algorithms obtain reductions in execution time with no change in performance. 1.

The best systems for machine translation of natural language are based on statistical models learned from data. Conventional representation of a statistical translation model requires substantial offline computation and representation in main memory. Therefore, the principal bottlenecks to the amount of data we can exploit and the complexity of models we can use are available memory and CPU time, and current state of the art already pushes these limits. With data size and model complexity continually increasing, a scalable solution to this problem is central to future improvement. Callison-Burch et al. (2005) and Zhang and Vogel (2005) proposed a solution that we call translation by pattern matching, which we bring to fruition in this dissertation. The training data itself serves as a proxy to the model; rules and parameters are computed on demand. It achieves our desiderata of minimal offline computation and compact representation, but is dependent on fast pattern matching algorithms on text. They demonstrated its application to a common model based on the translation of contiguous substrings, but leave some open problems. Among these is a question: can this approach match the performance of conventional methods despite unavoidable differences that it induces in the model? We show how to answer this question affirmatively. The main

Cross-language information retrieval (CLIR) today is dominated by techniques that use token-to-token mappings from bilingual dictionaries. Yet, state-of-the-art statistical translation models (e.g., using Synchronous Context-Free Grammars) are far richer, capturing multi-term phrases, term dependencies, and contextual constraints on translation choice. We present a novel CLIR framework that is able to reach inside the translation “black box ” and exploit these sources of evidence. Experiments on the TREC-5/6 English-Chinese test collection show this approach to be promising.

The small physical size of mobile devices imposes dramatic restrictions on the user interface (UI). With the ever increasing capacity of these devices as well as access to large online stores it becomes increasingly important to help the user select a particular item efficiently. Thus, we propose binary search with character pinning, where users can constrain their search to match selected prefix characters while making simple binary decisions about the position of their intended item in the lexicographic order. The underlying index for our method is based on a ternary search tree that is optimal under certain user-oriented constraints. To better scale to larger indexes, we analyze several heuristics that rapidly construct good trees. A user study demonstrates that our method helps users conduct rapid searches, using less keystrokes, compared to other methods. able to browse huge catalogs of recorded music. In order to address the market demand for tiny media players, different companies have introduced a wide range of small form factor products, with and without a small graphical user display (e.g. Figure 1). However, the small physical size of such devices only allows for a limited user interface (UI), typically a small number of buttons and a single line display. For example, the iPod Shuffle 1 has a directional pad (d-pad) with one center and four direction buttons, and no display unit. Given such a limited UI, selecting a song or artist from a large list (which could potentially be all the music available in an online store), becomes challenging. Simple approaches, such as iterating through the index item by item, fail to scale to large indexes.

This paper describes the system we developed to improve German-English translation of News text for the shared task of the Fifth Workshop on Statistical Machine Translation. Working within cdec, an open source modular framework for machine translation, we explore the benefits of several modifications to our hierarchical phrase-based model, including segmentation lattices, minimum Bayes Risk decoding, grammar extraction methods, and varying language models. Furthermore, we analyze decoder speed and memory performance across our set of models and show there is an important trade-off that needs to be made. 1

We describe the progress we have made in the past year on Joshua (Li et al., 2009a), an open source toolkit for parsing based machine translation. The new functionality includes: support for translation grammars with a rich set of syntactic nonterminals, the ability for external modules to posit constraints on how spans in the input sentence should be translated, lattice parsing for dealing with input uncertainty, a semiring framework that provides a unified way of doing various dynamic programming calculations, variational decoding for approximating the intractable MAP decoding, hypergraph-based discriminative training for better feature engineering, a parallelized MERT module, documentlevel and tail-based MERT, visualization of the derivation trees, and a cleaner pipeline for MT experiments. 1

We present a scalable discriminative training framework for parsing-based statistical machine translation. Our framework exploits hypergraphs (or packed forests) to compactly encode exponentially many competing translations, and uses the perceptron algorithm to learn to discriminatively prefer the oracle-best tree in the hypergraph. To facilitate training, we present: (i) an oracle extraction algorithm to efficiently extract the oracle trees from a hypergraph that best match the reference translations; (ii) a hypergraph pruning algorithm that substantially reduces the disk space required for storing the hypergraphs without degrading the translation quality; and (iii) simple yet effective data and feature selection algorithms by which an equally good or better model is obtained using a fraction of the training data and features. We experimentally show that our approach is scalable and is able to improve over a full-scale state-of-the-art hierarchical machine translation system. 1

Proceedings of the Workshop

We provide a general algorithmic schema for translation rule extraction and show that several popular extraction methods (including phrase pair extraction, hierarchical phrase pair extraction, and GHKM extraction) can be viewed as specific instances of this schema. This work is primarily intended as a survey of the dominant extraction paradigms, in which we make explicit the close relationship between these approaches, and establish a language for future hybridizations. This facilitates a generic and extensible implementation of alignment-based extraction methods. 1

This paper presents the system we developed for the 2011 WMT Haitian Creole–English SMS featured translation task. Applying standard statistical machine translation methods to noisy real-world SMS data in a low-density language setting such as Haitian Creole poses a unique set of challenges, which we attempt to address in this work. Along with techniques to better exploit the limited available training data, we explore the benefits of several methods for alleviating the additional noise inherent in the SMS and transforming it to better suite the assumptions of our hierarchical phrase-based model system. We show that these methods lead to significant improvements in BLEU score over the baseline. 1