Example summaries for DUC’04 produced by RL (see Section 5.4) with a reward function learnt from 100 user interactions using (a) the BT, UNC method of Gao et al. (2018) and (b) our GPPL, IMP method. (c) is a model summary written by an expert. Each color indicates a particular news event or topic, showing where it occurs in each summary. Compared to (a), summary (b) covers more of the events discussed in the reference, (c).
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy. No content on this site may be used to train artificial intelligence systems without permission in writing from the MIT Press.