Masked Measurement Prediction: Learning to Jointly Predict Quantities and Units from Textual Context

TL;DR

This paper introduces Masked Measurement Prediction, a new task for models to jointly predict numerical values and units from text, along with a generative model, GeMM, to improve numeracy in language models.

Contribution

The paper proposes a novel task and a generative model for joint number and unit prediction, enhancing numerical reasoning capabilities of language models.

Findings

Traditional models underperform on joint number-unit tasks

GeMM outperforms baselines in measurement prediction

Pretraining with MMP improves numeracy in transformer models

Abstract

Physical measurements constitute a large portion of numbers in academic papers, engineering reports, and web tables. Current benchmarks fall short of properly evaluating numeracy of pretrained language models on measurements, hindering research on developing new methods and applying them to numerical tasks. To that end, we introduce a novel task, Masked Measurement Prediction (MMP), where a model learns to reconstruct a number together with its associated unit given masked text. MMP is useful for both training new numerically informed models as well as evaluating numeracy of existing systems. In order to address this task, we introduce a new Generative Masked Measurement (GeMM) model that jointly learns to predict numbers along with their units. We perform fine-grained analyses comparing our model with various ablations and baselines. We use linear probing of traditional pretrained transformer models (RoBERTa) to show that they significantly underperform jointly trained number-unit models, highlighting the difficulty of this new task and the benefits of our proposed pretraining approach. We hope this framework accelerates the progress towards building more robust numerical reasoning systems in the future.