Minimum Bayes-Risk Decoding for Statistical Machine Translation

By Shankar Kumar, William J. Byrne. 2004

518 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Edinburgh’s Phrase-based Machine Translation Systems for WMT-14 (2014, background): “…Os-borne, 2008; Chiang et al., 2009), a distortion limit of 6, maximum phrase-length of 5, 100-best translation options, Minimum Bayes Risk decoding (Kumar and Byrne, 2004), Cube Pruning (Huang and Chiang, 2007), with a stack-size of 1000 during tuning and 5000 during test and the…”
Determinantal Point Processes for Machine Learning (2012, methodology): “First proposed by Goel and Byrne [59] for automatic speech recognition, MBR decoding has also been used successfully for word alignment and machine translation [86, 87].”
Posterior calibration and exploratory analysis for natural language processing models (2015, methodology): “, 2013); often the goal is to marginalize over structures to calculate and minimize an expected loss function, as in minimum Bayes risk decoding (Goodman, 1996; Kumar and Byrne, 2004), or to perform joint inference between early and later stages of NLP analysis (e.”
A Simple, Fast Diverse Decoding Algorithm for Neural Generation (2016, methodology): “One serious issue with the beam search algorithm which has long been recognized (Macherey et al., 2008; Tromble et al., 2008; Kumar and Byrne, 2004) is lack of diversity: candidates often differ only by punctuation or minor morphological variations, with most of the words overlapping.”
EU-BRIDGE MT: Combined Machine Translation (2014, background): “, 2012), a distortion limit of 6, a maximum phrase length of 5, 100-best translation options, Minimum Bayes Risk decoding (Kumar and Byrne, 2004), cube pruning (Huang and Chiang, 2007), with a stack size of 1000 during tuning and 5000 during testing and the no-reordering-overpunctuation heuristic.”
Diffusion-LM Improves Controllable Text Generation (2022, methodology): “In these settings, we apply Minimum Bayes Risk (MBR) decoding [26] to aggregate a set of samples S drawn from the Diffusion-LM , and select the sample that achieves the minimum expected risk under a loss function L (e.”
On the Impact of Various Types of Noise on Neural Machine Translation (2018, background): “…phrase length, and count bin features (Blunsom and Osborne, 2008; Chiang et al., 2009), a maximum phrase-length of 5, compact phrase table (Junczys-Dowmunt, 2012) minimum Bayes risk decoding (Kumar and Byrne, 2004), cube pruning (Huang and Chiang, 2007), with a stack-size of 1000 during tuning.”
The Edinburgh/JHU Phrase-based Machine Translation Systems for WMT 2015 (2015, methodology): “In decoding we applied cube pruning (Huang and Chiang, 2007) with a stack size of 5000 (reduced to 1000 for tuning), Minimum Bayes Risk decoding (Kumar and Byrne, 2004), a maximum phrase length of 5, a distortion limit of 6, 100-best translation options and the no-reordering-over-punctuation…”
Integrating an Unsupervised Transliteration Model into Statistical Machine Translation (2014, background): “, 2012), a distortion limit of 6, 100-best translation options, MBR decoding (Kumar and Byrne, 2004), Cube Pruning (Huang and Chiang, 2007), and the no-reordering-overpunctuation heuristic.”
Automatically Assessing Machine Summary Content Without a Gold Standard (2013, methodology): “Such rescoring is implemented using a minimum Bayes risk technique (Kumar and Byrne 2004; Tromble et al. 2008).”
Investigating the Usefulness of Generalized Word Representations in SMT (2014, background): “…sparse lexical and domain features (Hasler et al., 2012), a distortion limit of 6, 100-best translation options, Minimum Bayes Risk decoding (Kumar and Byrne, 2004), Cube Pruning (Huang and Chiang, 2007), with a stack-size of 1000 during tuning and 5000 during test and the…”
Continuous diffusion for categorical data (2022, methodology): “Using a diﬀusion framework provides a natural way of approximating Minimum Bayes-Risk decoding (Eikema and Aziz, 2021; Kumar and Byrne, 2004) by sampling a number of hypotheses, each from a diﬀerent initial noise vector, and selecting the one minimizing in expectation a metric of interest, in our…”
The CMU-Avenue French-English Translation System (2012, methodology): “Our baseline translation system uses Viterbi decoding while our ﬁnal system uses segment-level Minimum Bayes-Risk decoding (Kumar and Byrne, 2004) over 500-best lists using 1 - BLEU as the loss function.”
N-gram Counts and Language Models from the Common Crawl (2014, unknown): “, 2012), a distortion limit of 6, 100-best translation options, Minimum Bayes Risk decoding (Kumar and Byrne, 2004), Cube Pruning (Huang and Chiang, 2007), with a stack-size of 1000 during tuning and 5000 during test and the no-reordering-over-punctuation heuristic.”
Generalized Minimum Bayes Risk System Combination (2011, methodology): “MBR arose in Bayes decision theory (Duda et al., 2000) and has since been applied to speech recognition (Goel and Byrne, 2000) and machine translation (Kumar and Byrne, 2004).”
Edinburgh’s Statistical Machine Translation Systems for WMT16 (2016, methodology): “In decoding we applied cube pruning (Huang and Chiang, 2007) with a stack size of 5000 (reduced to 1000 for tuning), Minimum Bayes Risk decoding (Kumar and Byrne, 2004), a maximum phrase length of 5, a distortion limit of 6, 100-best translation options and the no-reordering-over-punctuation…”
Transformers are Meta-Reinforcement Learners (2022, background): “Here, we define consensus representation as the memory representation that is closest on average to all likely representations (Kumar & Byrne, 2004), i.e., minimizes the Bayes risk considering the set of episodic memories.”
Lattice-based Minimum Error Rate Training for Statistical Machine Translation (2008, methodology): “Translation hypotheses produced on the blind test data were reranked using the Minimum-Bayes Risk (MBR) decision rule (Kumar and Byrne, 2004; Tromble et al., 2008).”
Mutual Information and Diverse Decoding Improve Neural Machine Translation (2016, background): “…translations in phrase-based MT, including compact representations like lattices and hypergraphs (Macherey et al., 2008; Tromble et al., 2008; Kumar and Byrne, 2004), “traits” like translation length (Devlin and Matsoukas, 2012), bag-ging/boosting (Xiao et al., 2013), or multiple systems…”