Playing by the Book: An Interactive Game Approach for Action Graph Extraction from Text

Abstract

Understanding procedural text requires tracking entities, actions and effects as the narrative unfolds. We focus on the challenging real-world problem of action-graph extraction from material science papers, where language is highly specialized and data annotation is expensive and scarce. We propose a novel approach, Text2Quest, where procedural text is interpreted as instructions for an interactive game. A learning agent completes the game by executing the procedure correctly in a text-based simulated lab environment. The framework can complement existing approaches and enables richer forms of learning compared to static texts. We discuss potential limitations and advantages of the approach, and release a prototype proof-of-concept, hoping to encourage research in this direction.

arXiv commentAccepted to NAACL 2019 ESSP workshop (https://scientific-knowledge.github.io/)

References in corpus (7)

• Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding
• Learn What Not to Learn: Action Elimination with Deep Reinforcement Learning
• TextWorld: A Learning Environment for Text-based Games
• Question Answering as Global Reasoning over Semantic Abstractions
• Building Dynamic Knowledge Graphs from Text using Machine Reading Comprehension
• Deep Dyna-Q: Integrating Planning for Task-Completion Dialogue Policy Learning
• Automatically Extracting Action Graphs from Materials Science Synthesis Procedures