BriefGPT.xyz
Jun, 2017
多模态指令中的符号基础
Grounding Symbols in Multi-Modal Instructions
HTML
PDF
Yordan Hristov, Svetlin Penkov, Alex Lascarides, Subramanian Ramamoorthy
TL;DR
本文提出了一种方法来处理跨模态输入的原始流，以产生物体的细分并与高级概念相关联，以学习用户的颜色和形状的概念，并表明该模型可以从少量的物理演示中推广到识别新单词的物理指示。
Abstract
As
robots
begin to cohabit with humans in semi-structured environments, the need arises to understand
instructions
involving rich variability---for instance, learning to ground symbols in the physical world. Real
→