MemSum: Extractive Summarization of Long Documents Using Multi-Step Episodic Markov Decision Processes
This work addresses the challenge of generating high-quality, low-redundancy summaries for long documents, which is important for researchers and professionals dealing with large text corpora, though it is an incremental improvement over existing methods.
The paper tackled the problem of extractive summarization for long documents by introducing MemSum, a reinforcement learning model that considers sentence content, global context, and extraction history, achieving state-of-the-art ROUGE scores on PubMed, arXiv, and GovReport datasets.
We introduce MemSum (Multi-step Episodic Markov decision process extractive SUMmarizer), a reinforcement-learning-based extractive summarizer enriched at each step with information on the current extraction history. When MemSum iteratively selects sentences into the summary, it considers a broad information set that would intuitively also be used by humans in this task: 1) the text content of the sentence, 2) the global text context of the rest of the document, and 3) the extraction history consisting of the set of sentences that have already been extracted. With a lightweight architecture, MemSum obtains state-of-the-art test-set performance (ROUGE) in summarizing long documents taken from PubMed, arXiv, and GovReport. Ablation studies demonstrate the importance of local, global, and history information. A human evaluation confirms the high quality and low redundancy of the generated summaries, stemming from MemSum's awareness of extraction history.