Regenerating Hypotheses for Statistical Machine Translation

This paper studies three techniques that improve the quality of N-best hypotheses through additional regeneration process. Unlike the multi-system consensus approach where multiple translation systems are used, our improvement is achieved through the expansion of the N-best hypotheses from a single system. We explore three different methods to implement the regeneration process: redecoding, n-gram expansion, and confusion network-based regeneration. Experiments on Chinese-to-English NIST and IWSLT tasks show that all three methods obtain consistent improvements. Moreover, the combination of the three strategies achieves further improvements and outperforms the baseline by 0.81 BLEU-score on IWSLT'06, 0.57 on NIST'03, 0.61 on NIST'05 test set respectively.

• Publication year

  2008

• Venue

  International Conference on Computational Linguistics

• Publication date

  2008-08-18

• Fields of study

  Linguistics, Computer Science

• Identifiers
  DOI 10.3115/1599081.1599095
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Moses: Open Source Toolkit for Statistical Machine Translation

  2007cited by this paper
• Consensus Network Decoding for Statistical Machine Translation System Combination

  2007cited by this paper
• Combining Outputs from Multiple Machine Translation Systems

  2007influential reference
• Better n-best translations through generative n-gram language models

  2007influential reference
• Improved Word-Level System Combination for Machine Translation

  2007influential reference
• N-Gram Posterior Probabilities for Statistical Machine Translation

  2006influential reference
• Computing Consensus Translation for Multiple Machine Translation Systems Using Enhanced Hypothesis Alignment

  2006influential reference
• Reordering rules for phrase-based statistical machine translation

  2006cited by this paper
• Combining clues for lexical level aligning using the Null hypothesis approach

  2004cited by this paper
• Statistical Phrase-Based Translation

  2003cited by this paper
• Minimum Error Rate Training in Statistical Machine Translation

  2003cited by this paper
• A Systematic Comparison of Various Statistical Alignment Models

  2003influential reference
• Bleu: a Method for Automatic Evaluation of Machine Translation

  2002influential reference
• Toward a Broad-coverage Bilingual Corpus for Speech Translation of Travel Conversations in the Real World

  2002influential reference
• SRILM - an extensible language modeling toolkit

  2002cited by this paper
• Computing consensus translation from multiple machine translation systems

  2001cited by this paper
• Models of translation equivalence among words

  2000cited by this paper
• An Empirical Study of Smoothing Techniques for Language Modeling

  1996cited by this paper
• The Mathematics of Statistical Machine Translation: Parameter Estimation

  1993cited by this paper
• The mathematics of statistical machine translation

  1993cited by this paper

• Rewriter-Evaluator Architecture for Neural Machine Translation

  2020cites this paper
• Exploitation d'informations riches pour guider la traduction automatique statistique. (Complex Feature Guidance for Statistical Machine Translation)

  2016cites this paper
• A hybrid system for Chinese-English patent machine translation

  2015cites this paper
• Multi-Pass Decoding With Complex Feature Guidance for Statistical Machine Translation

  2015cites this paper
• Investigating the effects of controlled language on the reading and comprehension of machine translated texts: A mixed-methods approach

  2012cites this paper
• Fast Consensus Hypothesis Regeneration for Machine Translation

  2010influential citation
• Two-Stage Hypotheses Generation for Spoken Language Translation

  2009cites this paper
• A Comparative Study of Hypothesis Alignment and its Improvement for Machine Translation System Combination

  2009cites this paper
• I2R multi-pass machine translation system for IWSLT 2008.

  2008cites this paper