How to Train Your Agent to Read and Write

TL;DR

This paper introduces the DRAW network, an innovative system that enables an agent to read, understand, and generate scholarly text by extracting knowledge graphs, writing new content, and reviewing it for quality, advancing AI's ability to assist research tasks.

Contribution

The paper presents a novel Deep ReAder-Writer (DRAW) network that integrates knowledge extraction, text generation, and review, improving AI capabilities in reading and writing research papers.

Findings

DRAW outperforms baseline models on AGENDA datasets.

The system effectively extracts knowledge graphs from text.

It generates coherent and relevant paragraphs based on knowledge.

Abstract

Reading and writing research papers is one of the most privileged abilities that a qualified researcher should master. However, it is difficult for new researchers (\eg{students}) to fully {grasp} this ability. It would be fascinating if we could train an intelligent agent to help people read and summarize papers, and perhaps even discover and exploit the potential knowledge clues to write novel papers. Although there have been existing works focusing on summarizing (\emph{i.e.}, reading) the knowledge in a given text or generating (\emph{i.e.}, writing) a text based on the given knowledge, the ability of simultaneously reading and writing is still under development. Typically, this requires an agent to fully understand the knowledge from the given text materials and generate correct and fluent novel paragraphs, which is very challenging in practice. In this paper, we propose a Deep ReAder-Writer (DRAW) network, which consists of a \textit{Reader} that can extract knowledge graphs (KGs) from input paragraphs and discover potential knowledge, a graph-to-text \textit{Writer} that generates a novel paragraph, and a \textit{Reviewer} that reviews the generated paragraph from three different aspects. Extensive experiments show that our DRAW network outperforms considered baselines and several state-of-the-art methods on AGENDA and M-AGENDA datasets. Our code and supplementary are released at https://github.com/menggehe/DRAW.