SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.06264 · ACL · 2020

Tangled up in BLEU: Reevaluating the Evaluation of Automatic Machine Translation Evaluation Metrics

Timothy Baldwin, Trevor Cohn, Nitika Mathur

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Automatic metrics are fundamental for the development and evaluation of machine translation systems. Judging whether, and to what extent, automatic metrics concur with the gold standard of human evaluation is not a straightforward problem. We show that current methods for judging metrics are highly sensitive to the translations used for assessment, particularly the presence of outliers, which often leads to falsely confident conclusions about a metric's efficacy. Finally, we turn to pairwise system ranking, developing a method for thresholding performance improvement under an automatic metric against human judgements, which allows quantification of type I versus type II errors incurred, i.e., insignificant human differences in system quality that are accepted, and significant human differences that are rejected. Together, these findings suggest improvements to the protocols for metric evaluation and system performance evaluation in machine translation.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2006.06264")
get_code_for_paper("2006.06264")
have("2006.06264")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2006.06264.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2006.06264.svg)](https://syntology.ai/paper/2006.06264)

Connect an agent — have() is free.