The dominant paradigm for neural text generation is left-to-right decoding from
autoregressive language models. Constrained or controllable generation under complex …

We take a multi-pass approach to machine translation decoding when using synchronous
context-free grammars as the translation model and n-gram language models: the first pass …

We present a discriminative model that directly predicts which set of phrasal translation rules
should be extracted from a sentence pair. Our model scores extraction sets: nested …

This chapter introduces the basic concept and architecture of SMT and the various linguistic
challenges for SMT, which are the reasons that we introduce linguistically motivated SMT …

The translation of pragmatic meaning requires the processing of pragmatic implicit meaning,
the translator's responsibility for the readers of the target language to understand the …

We present the first known direct measurement of word alignment coverage on an Arabic-
English parallel corpus using inversion transduction grammar constraints. While direct …

We present a novel method for creating A* estimates for structured search problems. In our
approach, we project a complex model onto multiple simpler models for which exact …

Inversion transduction grammar (ITG) provides a syntactically motivated solution to modeling
the distortion of words between two languages. Although the Viterbi ITG alignments can be …

We present a novel method for creating A* estimates for structured search problems
originally described in Haghighi, DeNero, & Klein (2007). In our approach, we project a …

Statistical Machine Translation (SMT) produces results that make it apreferred choice in
most machine-assisted translation scenarios. However, the development of such high …