Atomic: An Atlas of Machine Commonsense for If-Then Reasoning
Maarten Sap†⋆ Ronan Le Bras† Emily Allaway⋆ Chandra Bhagavatula† Nicholas Lourie† Hannah Rashkin⋆ Brendan Roof† Noah A. Smith†⋆ Yejin Choi†⋆ ⋆Paul G. Allen School of Computer Science、Engineering, University of Washington, Seattle, USA †Allen Institute for Artificial Intelligence, Seattle, USA
Abstract
We present Atomic, an atlas of everyday commonsense reasoning, organized through 877k textual descriptions of inferential knowledge. Compared to existing resources that center around taxonomic knowledge, Atomic focuses on inferential knowledge organized as typed if-then relations with variables (e.g., “if X pays Y a compliment, then Y will likely return the compliment”). We propose nine if-then relation types to distinguish causes vs. effects, agents vs. themes, voluntary vs. involuntary events, and actions vs. mental states. By generatively training on the rich inferential knowledge described in Atomic, we show that neural models can acquire simple commonsense capabilities and reason about previously unseen events. Experimental results demonstrate that multitask models that incorporate the hierarchical structure of if-then relation types lead to more accurate inference compared to models trained in isolation, as measured by both automatic and human evaluation.
中文速览
人类天生会对日常事件进行因果推理——看到某人挡住了攻击，就能猜出他事先接受过训练、当时想要自保、事后会报警——但今天的AI系统普遍缺乏这种能力。为此，研究团队构建了Atomic，一个包含87.7万条文本的常识推理知识图谱，核心是九种"如果…那么…"关系类型，覆盖事件的原因、结果、心理状态和人物特征等维度，所有数据通过众包自由文本标注收集，而非从语料库中自动抽取，以避免语言报道偏差。基于Atomic训练的神经网络编解码模型能够对从未见过的新事件生成合理的因果推断，而利用九种关系层级结构设计的多任务模型在自动评估和人工评估上均显著优于独立训练的单任务模型。Atomic的价值在于为AI提供了一个系统化、大规模的"事件因果常识库"，是让机器真正理解日常世界运作规律的重要基础资源。
原文 arXiv:1811.00146;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1811.00146v3