"John is 50 years old, can his son be 65?" Evaluating NLP Models' Understanding of Feasibility

TL;DR

This paper introduces FeasibilityQA, a dataset to evaluate NLP models' understanding of action feasibility, revealing that even advanced models like GPT-3 struggle significantly with this commonsense reasoning task.

Contribution

The creation of FeasibilityQA dataset and comprehensive evaluation of state-of-the-art models' limitations in understanding action feasibility.

Findings

GPT-3 achieves only 19-62% accuracy on feasibility questions.

Additional knowledge improves performance by 7%.

Models show limited reasoning about action feasibility.

Abstract

In current NLP research, large-scale language models and their abilities are widely being discussed. Some recent works have also found notable failures of these models. Often these failure examples involve complex reasoning abilities. This work focuses on a simple commonsense ability, reasoning about when an action (or its effect) is feasible. To this end, we introduce FeasibilityQA, a question-answering dataset involving binary classification (BCQ) and multi-choice multi-correct questions (MCQ) that test understanding of feasibility. We show that even state-of-the-art models such as GPT-3, GPT-2, and T5 struggle to answer the feasibility questions correctly. Specifically, on MCQ and BCQ questions, GPT-3 achieves an accuracy of just (19%, 62%) and (25%, 64%) in zero-shot and few-shot settings, respectively. We also evaluate models by providing relevant knowledge statements required to answer the question. We find that the additional knowledge leads to a 7% gain in performance, but the overall performance still remains low. These results make one wonder how much commonsense knowledge about action feasibility is encoded in state-of-the-art models and how well they can reason about it.