PoBRL: Optimizing Multi-Document Summarization by Blending Reinforcement Learning Policies
This addresses the problem of generating high-quality summaries from multiple documents, which is incremental as it builds on existing RL methods for summarization.
The paper tackled multi-document summarization by proposing PoBRL, a reinforcement learning framework that jointly optimizes importance, relevance, and length, achieving state-of-the-art performance on several datasets.
We propose a novel reinforcement learning based framework PoBRL for solving multi-document summarization. PoBRL jointly optimizes over the following three objectives necessary for a high-quality summary: importance, relevance, and length. Our strategy decouples this multi-objective optimization into different subproblems that can be solved individually by reinforcement learning. Utilizing PoBRL, we then blend each learned policies together to produce a summary that is a concise and complete representation of the original input. Our empirical analysis shows state-of-the-art performance on several multi-document datasets. Human evaluation also shows that our method produces high-quality output.