NumGPT: Improving Numeracy Ability of Generative Pre-trained Models
This addresses the issue of poor numerical reasoning in AI models for applications requiring accurate number handling, but it is incremental as it builds on existing GPT architectures.
The paper tackles the problem of generative pre-trained language models lacking robust numeracy abilities by proposing NumGPT, which explicitly models numerical properties in texts, resulting in outperforming baseline models like GPT on tasks such as measurement estimation and math word problems.
Existing generative pre-trained language models (e.g., GPT) focus on modeling the language structure and semantics of general texts. However, those models do not consider the numerical properties of numbers and cannot perform robustly on numerical reasoning tasks (e.g., math word problems and measurement estimation). In this paper, we propose NumGPT, a generative pre-trained model that explicitly models the numerical properties of numbers in texts. Specifically, it leverages a prototype-based numeral embedding to encode the mantissa of the number and an individual embedding to encode the exponent of the number. A numeral-aware loss function is designed to integrate numerals into the pre-training objective of NumGPT. We conduct extensive experiments on four different datasets to evaluate the numeracy ability of NumGPT. The experiment results show that NumGPT outperforms baseline models (e.g., GPT and GPT with DICE) on a range of numerical reasoning tasks such as measurement estimation, number comparison, math word problems, and magnitude classification. Ablation studies are also conducted to evaluate the impact of pre-training and model hyperparameters on the performance.