Skip to yearly menu bar Skip to main content


Model Comparison for Semantic Grouping

Francisco Vargas · Kamen Brestnichki · Nils Hammerla

Pacific Ballroom #219

Keywords: [ Unsupervised Learning ] [ Natural Language Processing ] [ Metric Learning ] [ Interpretability ] [ Information Retrieval ]


We introduce a probabilistic framework for quantifying the semantic similarity between two groups of embeddings. We formulate the task of semantic similarity as a model comparison task in which we contrast a generative model which jointly models two sentences versus one that does not. We illustrate how this framework can be used for the Semantic Textual Similarity tasks using clear assumptions about how the embeddings of words are generated. We apply model comparison that utilises information criteria to address some of the shortcomings of Bayesian model comparison, whilst still penalising model complexity. We achieve competitive results by applying the proposed framework with an appropriate choice of likelihood on the STS datasets.

Live content is unavailable. Log in and register to view live content