The Dangers of Underclaiming: Reasons for Caution When Reporting How NLP Systems Fail

TL;DR

This paper highlights the risks of underreporting NLP system failures, emphasizing the importance of accurate claims to maintain credibility and effectively address current and future challenges in the field.

Contribution

It advocates for more careful reporting of NLP system limitations and proposes strategies to improve communication and prevent misleading claims.

Findings

Underreporting failures harms credibility and trust.

Accurate failure reporting can improve system development and public understanding.

Strategies for better communication of NLP limitations are proposed.

Abstract

Researchers in NLP often frame and discuss research results in ways that serve to deemphasize the field's successes, often in response to the field's widespread hype. Though well-meaning, this has yielded many misleading or false claims about the limits of our best technology. This is a problem, and it may be more serious than it looks: It harms our credibility in ways that can make it harder to mitigate present-day harms, like those involving biased systems for content moderation or resume screening. It also limits our ability to prepare for the potentially enormous impacts of more distant future advances. This paper urges researchers to be careful about these claims and suggests some research directions and communication strategies that will make it easier to avoid or rebut them.