Sparse Optimization for Unsupervised Extractive Summarization of Long Documents with the Frank-Wolfe Algorithm
This provides an efficient unsupervised method for summarizing long documents, which is incremental but offers practical improvements.
The authors tackled unsupervised extractive summarization of long documents by modeling it as a sparse auto-regression problem and solving it with a Frank-Wolfe algorithm that requires only about k iterations for a k-sentence summary. Their method outperformed two other unsupervised approaches on both lexical and semantic metrics, particularly when combined with embeddings for paraphrased summaries.
We address the problem of unsupervised extractive document summarization, especially for long documents. We model the unsupervised problem as a sparse auto-regression one and approximate the resulting combinatorial problem via a convex, norm-constrained problem. We solve it using a dedicated Frank-Wolfe algorithm. To generate a summary with $k$ sentences, the algorithm only needs to execute $\approx k$ iterations, making it very efficient. We explain how to avoid explicit calculation of the full gradient and how to include sentence embedding information. We evaluate our approach against two other unsupervised methods using both lexical (standard) ROUGE scores, as well as semantic (embedding-based) ones. Our method achieves better results with both datasets and works especially well when combined with embeddings for highly paraphrased summaries.