Deriving Commonsense Inference Tasks from Interactive Fictions
arXiv:2010.09788
Abstract
Commonsense reasoning simulates the human ability to make presumptions about our physical world, and it is an indispensable cornerstone in building general AI systems. We propose a new commonsense reasoning dataset based on human's interactive fiction game playings as human players demonstrate plentiful and diverse commonsense reasoning. The new dataset mitigates several limitations of the prior art. Experiments show that our task is solvable to human experts with sufficient commonsense knowledge but poses challenges to existing machine reading models, with a big performance gap of more than 30%.
References in corpus (7)
- Machine Comprehension Using Match-LSTM and Answer Pointer
- Graph Constrained Reinforcement Learning for Natural Language Action Spaces
- Abductive Commonsense Reasoning
- Interactive Fiction Games: A Colossal Adventure
- How to Avoid Being Eaten by a Grue: Structured Exploration Strategies for Textual Worlds
- WordCraft: An Environment for Benchmarking Commonsense Agents
- Interactive Fiction Game Playing as Multi-Paragraph Reading Comprehension with Reinforcement Learning