arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2401.16247cs.CLcs.CY
收藏

面向多模态与多语言翻译中的红队测试研究

Towards Red Teaming in Multimodal and Multilingual Translation

FAIR, Meta(Meta AI 基础人工智能研究中心)
Christophe Ropers, David Dale, Prangthip Hansanti, Gabriel Mejia Gonzalez, Ivan Evtimov, Corinne Wong, Christophe Touret, Kristina Pereyra, Seohyun Sonia Kim, C… 展开作者

Christophe Ropers, David Dale, Prangthip Hansanti, Gabriel Mejia Gonzalez, Ivan Evtimov, Corinne Wong, Christophe Touret, Kristina Pereyra, Seohyun Sonia Kim, Cristian Canton Ferrer, Pierre Andrews, Marta R. Costa-jussà

AI总结:

针对自然语言处理评估数据集易与训练数据重叠导致性能高估的问题,本文首次开展机器翻译领域的人工红队测试研究,探索人工与自动化红队测试方案,总结经验并提出建议,为该领域研究开辟新方向。

摘要:

自然语言处理领域的性能评估正变得日益复杂。一个尤为突出的挑战是评估数据集可能与训练数据存在直接或间接的重叠,这会导致结果出现偏差,高估模型性能。因此,人工评估作为衡量模型性能与可靠性的手段,正受到越来越多的关注。红队测试方法便是其中一种,其目标是生成能让模型产生严重错误的边缘案例。尽管该方法正逐渐成为生成式AI的标准实践,但它在条件式AI领域的应用仍大多未被探索。本文首次开展了针对机器翻译(Machine Translation,MT)的人工红队测试研究,朝着理解并提升翻译模型性能迈出了重要一步。我们深入研究了人工红队测试与自动化相关实验,总结了经验教训,并针对翻译模型和红队测试演练提出了建议。这项开创性工作为机器翻译领域的研究与开发开辟了新路径。

英文摘要:

Assessing performance in Natural Language Processing is becoming increasingly complex. One particular challenge is the potential for evaluation datasets to overlap with training data, either directly or indirectly, which can lead to skewed results and overestimation of model performance. As a consequence, human evaluation is gaining increasing interest as a means to assess the performance and reliability of models. One such method is the red teaming approach, which aims to generate edge cases where a model will produce critical errors. While this methodology is becoming standard practice for generative AI, its application to the realm of conditional AI remains largely unexplored. This paper presents the first study on human-based red teaming for Machine Translation (MT), marking a significant step towards understanding and improving the performance of translation models. We delve into both human-based red teaming and a study on automation, reporting lessons learned and providing recommendations for both translation models and red teaming drills. This pioneering work opens up new avenues for research and development in the field of MT.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑