Search Results

1 - 2 of 2 items

  • Author: Juan Pinos x
Clear All Modify Search

Abstract

Polylepis species represent one of the most important and endemic woodlands of the mid- and high-elevation regions of the Andean Cordillera. I provide a review of the current situation of Polylepis woodlands, discuss the potential effects of various conservation measures and consider the likely impact of climate change on tree phenology and tree regeneration, aiming to foster the conservation and sustainable management of these woodlands through proper environmental planning. I argue that in addition to the delineation and extension of protected areas, it is essential to incorporate actions such as forestation, forest policies, environmental education and local community participation. To be effective, conservation measures should be implemented in an international transdisciplinary research framework and in harmony with site-specific conditions. Finally, given the likely but uncertain influences of climate change on Polylepis woodlands, further research (and communication of that research) is needed to improve forest management strategies and research priorities for the Andean region.

Simple and Efficient Model Filtering in Statistical Machine Translation

Data availability and distributed computing techniques have allowed statistical machine translation (SMT) researchers to build larger models. However, decoders need to be able to retrieve information efficiently from these models to be able to translate an input sentence or a set of input sentences. We introduce an easy to implement and general purpose solution to tackle this problem: we store SMT models as a set of key-value pairs in an HFile. We apply this strategy to two specific tasks: test set hierarchical phrase-based rule filtering and n-gram count filtering for language model lattice rescoring. We compare our approach to alternative strategies and show that its trade offs in terms of speed, memory and simplicity are competitive.