PaperRobot: Incremental Draft Generation of Scientific Ideas

Qingyun Wang, Lifu Huang, Zhiying Jiang, Kevin Knight, Heng Ji +2 more
5/20/2019
cs.CLcs.AIcs.LG

Official code: EagleW/PaperRobot (479 stars, MIT, PyTorch, last commit 2024-03-09).

Abstract

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.

View on arXivView PDF

Code Implementations(1)

EagleW/PaperRobotOfficial100%
PyTorchMIT

Code for PaperRobot: Incremental Draft Generation of Scientific Ideas

479128PythonMay 18, 20192 years agoMIT
attention-mechanismdatasetsend-to-end-learninggenerationmemory-networks+5 more

Similar Papers

Graphs of Research: Citation Evolution Graphs as Supervision for Research Idea Generation

82% match2026

Paper Abstract Writing through Editing Mechanism

77% match2 repos742018

Citing Less Critically: LLMs Reshape the Rhetoric and Reach of Scientific Citation

76% match2026

ARIES: A Corpus of Scientific Paper Edits Made in Response to Peer Reviews

76% match1 repo562023

BioGPT: Generative Pre-trained Transformer for Biomedical Text Generation and Mining

76% match1 repo4.5k2022

Unfolding Scientific Papers into Multi-Turn Generation Trajectories for Continued Pre-Training

76% match2026

Cite this paper

@article{wang2019paperrobot,
  title  = {PaperRobot: Incremental Draft Generation of Scientific Ideas},
  author = {Qingyun Wang and Lifu Huang and Zhiying Jiang and Kevin Knight and Heng Ji and Mohit Bansal and Yi Luan},
  year   = {2019},
  eprint = {1905.07870},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1905.07870}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1905.07870.svg)](https://codeofpaper.com/paper/1905.07870)

Discussion