TensorX
返回文献探索

Paper · arXiv 2608.02589

CAPEval: A Decoupled Caption Evaluation across Understanding and Generation

Zhipeng Liu, Haochen Wang, Zhaoxiang Zhang

25 upvotesAugust 3, 2026arXiv 预印本
AI 摘要

CAPEval separates caption quality into coverage and precision, revealing that coverage predicts understanding performance while precision predicts generation performance.

CAPEvalcoverageprecisioncaption evaluationmultimodal understandingtext-to-image generationatomic checklist

Abstract

Captions serve as a primary supervision signal for both multimodal understanding and text-to-image generation. However, previous evaluations treat the caption quality as a single scalar objective, which conflates two distinct properties: (1) how much visual information a caption covers and (2) how reliably the image supports its stated claims. To this end, we design a decoupled caption evaluation benchmark, CAPEval (Coverage And Precision Evaluation), with human-written ground-truth captions and human-verified atomic checklist items. Specifically, CAPEval decomposes caption quality into Coverage and Precision. The former quantifies how thoroughly a caption covers ground-truth factual content, while the latter reflects the factual correctness rate of all claims expressed in the caption. We select 10 captioners and further conduct controlled downstream end-to-end experiments with them from four model families, where the caption source is the only variable. Empirically, we find a consistent task-dependent dissociation: Coverage serves as the stronger correlate for understanding performance, whereas Precision acts as the dominant predictor for generation performance. This decoupled evaluation paradigm not only delivers a more fine-grained diagnosis of caption quality, but also offers actionable guidance for selecting and optimizing captioners tailored to different downstream tasks.

北京市昌平区探索星信息技术及软件开发工作室

京ICP备2026059466号
CAPEval: A Decoupled Caption Evaluation across Understanding and Generation | TensorX