Calibrate your listeners! Robust communication-based training for pragmatic speakers
Rose Wang, Julia A. White, Jesse Mu, Noah D. Goodman · 2021
To be good conversational partners, natural language processing (NLP) systems should be trained to produce contextually useful utterances.Prior work has investigated training NLP systems with communication-based objectives, where a neural listener stands in as a communication partner.However, these systems commonly suffer from semantic drift where the learned language diverges radically from natural language.We propose a method that uses a population of neural listeners to regularize speaker training.We first show that language drift originates from the poor uncertainty calibration of a neural listener, which makes high-certainty predictions on novel sentences.We explore ensemble-and dropoutbased populations of listeners and find that the former results in better uncertainty quantification.We evaluate both population-based objectives on reference games, and show that the ensemble method with better calibration enables the speaker to generate pragmatic utterances while scaling to a large vocabulary and generalizing to new games and listeners.1