arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.18756cs.CL
收藏

比较摘要模型需要多少标注?

How Much Annotation is Needed to Compare Summarization Models?

Northeastern University(东北大学);Pattern Data(帕特恩数据公司);Adobe Research(奥多比研究院)

Chantal Shaib, Joe Barrow, Alexa F. Siu, Byron C. Wallace, Ani Nenkova

AI总结:

本文实证研究新闻摘要模型选择所需的测试样本量,发现自动与人工比较评估在不足100个样本时即可收敛,并评估了自动指标预测人工偏好胜率的能力。

摘要:

现代指令微调模型在摘要等文本生成任务上已具备很强能力,并且预计将持续稳定发布。在实践中,人们现在可能希望在将模型应用于新领域或新目的时,以最小的成本自信地选出性能最佳的摘要模型。在本研究中,我们通过实证考察了在新闻摘要场景下选择偏好模型所需的测试样本量。实证结果表明,自动评估和人工评估的比较评价都会快速收敛,仅用不到100个样本就能形成对某个系统的明确偏好。人工偏好数据使我们能够量化自动评分在多种下游摘要任务中复现偏好排序的能力。我们发现,尽管自动指标在较小样本量下表现稳定,但只有部分自动指标能够根据人工偏好对模型胜率作出中等程度的预测。

英文摘要:

Modern instruction-tuned models have become highly capable in text generation tasks such as summarization, and are expected to be released at a steady pace. In practice one may now wish to choose confidently, but with minimal effort, the best performing summarization model when applied to a new domain or purpose. In this work, we empirically investigate the test sample size necessary to select a preferred model in the context of news summarization. Empirical results reveal that comparative evaluation converges quickly for both automatic and human evaluation, with clear preferences for a system emerging from under 100 examples. The human preference data allows us to quantify how well automatic scores can reproduce preference rankings across a variety of downstream summarization tasks. We find that, while automatic metrics are stable at smaller sample sizes, only some automatic metrics are able to moderately predict model win rates according to human preference.

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。

↑