"John is 50 years old, can his son be 65?" Evaluating NLP Models' Understanding of Feasibility
This work addresses a critical gap in assessing commonsense reasoning for NLP researchers, though it is incremental as it builds on existing evaluation methods.
The paper tackles the problem of evaluating NLP models' understanding of feasibility in commonsense reasoning, showing that state-of-the-art models like GPT-3 achieve low accuracies, such as 19% to 64% on binary and multi-choice questions, with only a 7% gain from additional knowledge.
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.