Generalizing Translation Models in the Probabilistic Relevance Framework

By Navid Rekabsaz, Mihai Lupu, Allan Hanbury, Guido Zuccon. 2016

26 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 Artificial Intelligence

Citations in the corpus

Relevance-based Word Embedding (2017, background): “…models have been demonstrated to be effective in a number of IR tasks, including query expansion [11, 17, 40], query classification [23, 41], short text similarity [15], and document model estimation [2, 31]. aforementioned embedding models are typically trained basedontermproximityinalarge corpus.”
Neural Ranking Models with Weak Supervision (2017, methodology): “ng these, distributed word representations or embeddings [27, 31] have a−racted a lot of a−ention. Word embedding vectors have been applied to term re-weighting in IR models [32, 47], query expansion [10, 33, 43], query classi•cation [25, 44], etc. „e main shortcomingofmostoftheapproachesinthiscateg”
Word Embedding Causes Topic Shifting; Exploit Global Context! (2017, methodology): “Zuccon et al. [18] integrate word embedding with the translation language modeling by using the set of extended terms from word embedding: Rekabsaz et al. [12] extend the idea into four probabilistic relevance frameworks.”
On the Effect of Low-Frequency Terms on Neural-IR Models (2019, methodology): “Various studies apply neural methods by introducing pre-trained word embeddings into classical IR models [13, 14],”
A Modern Perspective on Query Likelihood with Deep Generative Retrieval Models (2021, unknown): “This is done for instance by generalizing term salience with translation models [38, 40], and re-weighting terms [57], or through adapting word embeddings for document retrieval by post-filtering [39], retrofitting [16], or re-training on local documents [9].”
Neural information retrieval: at the end of the early years (2017, methodology): “For the Implicit approach, the term similarity function is a design choice whereas the TTU similarity function stands as a design choice for the Explicit category.”