Interpolated estimation of Markov source parameters from sparse data

By Frederick Jelinek. 1980

Subfields: Probabilistic & Bayesian Methods

How researchers see this paper

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

In Probabilistic & Bayesian Methods

Citations in the corpus

Learning Visual N-Grams from Web Data (2016, methodology): “This loss uses smoothing to assign non-zero probabilities to out-of-dictionary n-grams; specifically, we experiment with Jelinek-Mercer smoothing [26].”
A bit of progress in language modeling (2001, methodology): “While the most commonly used smoothing techniques, Katz smoothing (Katz, 1987) and Jelinek-Mercer smoothing (Jelinek and Mercer, 1980) (sometimes called deleted interpolation) work fine, even better smoothing techniques exist.”
Exponentially Increasing the Capacity-to-Computation Ratio for Conditional Computation in Deep Learning (2014, methodology): “Beside the obvious regularization of choosing small k , we propose here an additional device that is inspired by the impressive success of smoothed or interpolated n-grams and back-off models in statistical language modeling (Katz, 1987; Jelinek and Mercer, 1980).”
Query Expansion Techniques for Information Retrieval: a Survey (2017, unknown): “Hence, reference [58] suggested that instead of considering a particular method, one can come up with a superior expanded query model (calculated using Jelinek-Mercer interpolation [129]) given as:”
Speech repains, intonational phrases, and discourse markers: modeling speakers’ utterances in spoken dialogue (1999, methodology): “Two common approaches for doing this are interpolated estimation (Jelinek and Mercer, 1980) and the backoff approach (Katz, 1987).”
Neural Network Methods for Natural Language Processing (2017, unknown): “7[Johnson and Zhang, 2015] 8[Wang et al., 2015a] 9[Kalchbrenner et al., 2014, Kim, 2014] 10[dos Santos et al., 2015, Zeng et al., 2014] 11[Chen et al., 2015, Nguyen and Grishman, 2015] 12[Yin and Schütze, 2015] 13[Collobert et al., 2011] 14[Dong et al., 2015] 15[Bitvai and Cohn, 2015] 16[Gao et al.,”
Language Models are Realistic Tabular Data Generators (2022, background): “Commonly, the probability of natural-language sequences is factorized in an auto-regressive manner in LLMs (Jelinek, 1980; Bengio et al., 2000).”
Grapheme-to-phoneme conversion for Chinese text-to-speech (2004, methodology): “To cope with data sparseness in MLE, linear interpolation technique [8] is used for word bigram smoothing.”
Visual Contextual Advertising: Bringing Textual Advertisements to Images (2010, methodology): “5 is set to be 0.15 as indicated in (Jelinek and Mercer 1980).”
Knowledge-Aware Language Model Pretraining (2020, background): “, wn}, the classical unidirectional factorization of language probabilities [22, 23] describes:”