Reproduce “An Efficient Coarse-to-Fine Facet-Aware Unsupervised Summarization Framework based on Semantic Blocks” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Xinnian Liang, Jing Li, Shuangzhi Wu, Jiali Zeng, Yufan Jiang, Mu Li, Zhoujun Li
C2F-FAR presents an unsupervised extractive method for summarizing long documents by ranking semantic blocks before sentences. It segments neighboring sentences into blocks, filters blocks by directed centrality, then selects and ranks sentences within the retained blocks. On Gov-Report, it scores 57.98 ROUGE-1 versus FAR’s 57.51 and reports 4–28× faster inference, but trails FAR on arXiv and PubMed.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.