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 2601.04609cs.CL
收藏

当更多词语说更少:在图像描述评估中解耦长度与特异性

When More Words Say Less: Decoupling Length and Specificity in Image Description Evaluation

Stanford University(斯坦福大学);University of California, Los Angeles(加州大学洛杉矶分校)

Rhea Kapur, Robert Hawkins, Elisa Kreiss

AI总结:

本文提出通过对比集定义描述特异性,验证人们更偏好信息密集的描述,无论长度如何,且长度分配影响特异性差异。

摘要:

视觉-语言模型(VLMs)正被越来越多地用于通过文本描述使视觉内容可访问。然而,在当前系统中,描述的特异性往往与长度混淆。我们主张必须将这两个概念解耦:描述可以简洁但信息密集,或冗长但空洞。我们定义特异性相对于对比集,其中描述的特异性程度在于其比其他可能图像更能识别目标图像。我们构建了一个控制长度而变化信息内容的数据集,并验证人们倾向于更特异的描述,无论长度如何。我们发现仅控制长度无法解释特异性差异:如何分配长度预算会产生影响。这些结果支持直接优先考虑特异性而非冗长的评估方法。

英文摘要:

Vision-language models (VLMs) are increasingly used to make visual content accessible via text-based descriptions. In current systems, however, description specificity is often conflated with their length. We argue that these two concepts must be disentangled: descriptions can be concise yet dense with information, or lengthy yet vacuous. We define specificity relative to a contrast set, where a description is more specific to the extent that it picks out the target image better than other possible images. We construct a dataset that controls for length while varying information content, and validate that people reliably prefer more specific descriptions regardless of length. We find that controlling for length alone cannot account for differences in specificity: how the length budget is allocated makes a difference. These results support evaluation approaches that directly prioritize specificity over verbosity.

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。

↑