A Better Variant of Self-Critical Sequence Training

03/22/2020

∙

by   Ruotian Luo, et al.

∙

0

∙

share

In this work, we present a simple yet better variant of Self-Critical Sequence Training. We make a simple change in the choice of baseline function in REINFORCE algorithm. The new baseline can bring better performance with no extra cost, compared to the greedy decoding baseline.