PaperRobot: Incremental Draft Generation of Scientific Ideas

We present a PaperRobot who performs as an automatic research assistant by (1) conducting deep understanding of a large collection of human-written papers in a target domain and constructing comprehensive background knowledge graphs (KGs); (2) creating new ideas by predicting links from the background KGs, by combining graph attention and contextual text attention; (3) incrementally writing some key elements of a new paper based on memory-attention networks: from the input title along with predicted related entities to generate a paper abstract, from the abstract to generate conclusion and future work, and finally from future work to generate a title for a follow-on paper. Turing Tests, where a biomedical domain expert is asked to compare a system output and a human-authored string, show PaperRobot generated abstracts, conclusion and future work sections, and new titles are chosen over human-written ones up to 30%, 24% and 12% of the time, respectively...

PDF Abstract ACL 2019 PDF ACL 2019 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Paper generation (abstract-to-conclusion) PubMed Term, Abstract, Conclusion, Title Dataset PaperRobot (-Repetition Removal) Meteor 12.3 # 1
Paper generation (Conclusion-to-title) PubMed Term, Abstract, Conclusion, Title Dataset PaperRobot Meteor 8.9 # 1
Paper generation (Title-to-abstract) PubMed Term, Abstract, Conclusion, Title Dataset PaperRobot METEOR 13 # 1

Methods