A Simple Machine Learning Method for Commonsense Reasoning? A Short Commentary on Trinh & Le (2018)

TL;DR

This commentary critiques Trinh & Le's simple ML approach for commonsense reasoning, highlighting flaws and arguing that data-driven models are insufficient for natural language understanding and reference resolution.

Contribution

It identifies critical flaws in a recent simple ML method for commonsense reasoning and discusses limitations of data-driven approaches in natural language understanding.

Findings

Highlights three serious flaws in Trinh & Le's method

Argues data-driven approaches are inadequate for commonsense reasoning

Emphasizes the complexity of natural language understanding

Abstract

This is a short Commentary on Trinh & Le (2018) ("A Simple Method for Commonsense Reasoning") that outlines three serious flaws in the cited paper and discusses why data-driven approaches cannot be considered as serious models for the commonsense reasoning needed in natural language understanding in general, and in reference resolution, in particular.