NumGLUE: A Suite of Fundamental yet Challenging Mathematical Reasoning Tasks
This addresses the problem of brittle mathematical reasoning in AI systems for researchers and developers, though it is incremental as it builds on existing benchmark concepts.
The authors introduced NumGLUE, a benchmark of eight mathematical reasoning tasks to evaluate AI systems' arithmetic understanding in text, showing that state-of-the-art models perform 46.4% worse than humans and that joint training across tasks yields an average 3.4% gain per task.
Given the ubiquitous nature of numbers in text, reasoning with numbers to perform simple calculations is an important skill of AI systems. While many datasets and models have been developed to this end, state-of-the-art AI systems are brittle; failing to perform the underlying mathematical reasoning when they appear in a slightly different scenario. Drawing inspiration from GLUE that was proposed in the context of natural language understanding, we propose NumGLUE, a multi-task benchmark that evaluates the performance of AI systems on eight different tasks, that at their core require simple arithmetic understanding. We show that this benchmark is far from being solved with neural models including state-of-the-art large-scale language models performing significantly worse than humans (lower by 46.4%). Further, NumGLUE promotes sharing knowledge across tasks, especially those with limited training data as evidenced by the superior performance (average gain of 3.4% on each task) when a model is jointly trained on all the tasks as opposed to task-specific modeling. Finally, we hope that NumGLUE will encourage systems that perform robust and general arithmetic reasoning within language, a first step towards being able to perform more complex mathematical reasoning.