arxivst stuff from arxiv that you should probably bookmark

Trainable Referring Expression Generation using Overspecification Preferences

Abstract · Apr 12, 2017 10:47 ·

overspecification speakers ball referential speaker descriptions tilburg reg personalised cs-cl

Arxiv Abstract

  • Thiago castro Ferreira
  • Ivandre Paraboni

Referring expression generation (REG) models that use speaker-dependent information require a considerable amount of training data produced by every individual speaker, or may otherwise perform poorly. In this work we present a simple REG experiment that allows the use of larger training data sets by grouping speakers according to their overspecification preferences. Intrinsic evaluation shows that this method generally outperforms the personalised method found in previous work.

Read the paper (pdf) »