Factorizing Content and Budget Decisions in Abstractive Summarization of Long Documents
This work addresses the challenge of generating concise summaries for long documents, offering improved performance and domain adaptation, though it is incremental as it builds on existing summarization methods.
The paper tackles the problem of abstractive summarization of long documents by disentangling content selection from budget allocation, resulting in significantly higher ROUGE scores on benchmarks like PubMed, arXiv, and GovReport, with a notable 46.29 ROUGE-1 score on arXiv when trained only on PubMed.
We argue that disentangling content selection from the budget used to cover salient content improves the performance and applicability of abstractive summarizers. Our method, FactorSum, does this disentanglement by factorizing summarization into two steps through an energy function: (1) generation of abstractive summary views; (2) combination of these views into a final summary, following a budget and content guidance. This guidance may come from different sources, including from an advisor model such as BART or BigBird, or in oracle mode -- from the reference. This factorization achieves significantly higher ROUGE scores on multiple benchmarks for long document summarization, namely PubMed, arXiv, and GovReport. Most notably, our model is effective for domain adaptation. When trained only on PubMed samples, it achieves a 46.29 ROUGE-1 score on arXiv, which indicates a strong performance due to more flexible budget adaptation and content selection less dependent on domain-specific textual structure.