FACE: Evaluating Natural Language Generation with Fourier Analysis of Cross-Entropy

Zuhao Yang, Yingfang Yuan, Yang Xu, SHUO ZHAN, Huajun Bai, Kefan Chen

Keywords: natural language generation; evaluation metrics; cross-entropy; language model

TL;DR: Fourier transform of cross-entropy sequences can be used to distinguish human from model-generated language.

Abstract: Measuring the distance between machine-produced and human language is a critical open problem. Inspired by empirical findings from psycholinguistics on the periodicity of entropy in language, we propose FACE, a set of metrics based on Fourier Analysis of the estimated Cross-Entropy of language, for measuring the similarity between model-generated and human-written languages. Based on an open-ended generation task and the experimental data from previous studies, we find that FACE can effectively identify the human-model gap, scales with model size, reflects the outcomes of different sampling methods for decoding, correlates well with other evaluation metrics and with human judgment scores.

Submission Number: 5583