# The price of debiasing automatic metrics in natural language evaluation

Source: [The price of debiasing automatic metrics in natural language evaluation](<https://www.sota2.com/research/paper/68fa64851e209131cea051b9>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T21:06:03.000Z

Authors: Arun Tejasvi Chaganty, Stephen Mussman, Percy Liang

Published: 2018-07-06T00:11:27.000Z

## Abstract

For evaluating generation systems, automatic metrics such as BLEU cost nothing to run but have been shown to correlate poorly with human judgment, leading to systematic bias against certain model improvements. On the other hand, averaging human judgments, the unbiased gold standard, is often too expensive. In this paper, we use control variates to combine automatic metrics with human evaluation to obtain an unbiased estimator with lower cost than human evaluation alone. In practice, however, we obtain only a 7-13% cost reduction on evaluating summarization and open-response question answering systems. We then prove that our estimator is optimal: there is no unbiased estimator with lower cost. Our theory further highlights the two fundamental bottlenecks---the automatic metric and the prompt shown to human evaluators---both of which need to be improved to obtain greater cost savings.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1807.02202>)

[PDF](<https://arxiv.org/pdf/1807.02202.pdf>)
