救赎分数：基于分布、感知与语言信号三角验证的图像描述多模态评估框架
Redemption Score: A Multi-Modal Evaluation Framework for Image Captioning via Distributional, Perceptual, and Linguistic Signal Triangulation
AI总结：
提出救赎分数（RS）框架，通过三角验证分布、感知与语言三种信号，无需任务训练即可在Flickr8k等基准上实现与人类判断高度相关的图像描述评估。
摘要：
评估图像描述需要对视觉语义和语言语用进行统一的评估，而这往往是大多数指标无法完全捕捉的。随着这些指标越来越多地指导多模态人工智能中的模型开发、基准测试和系统优化，评估中的不准确可能会歪曲真实的进展。我们引入了救赎分数（Redemption Score，RS），一种新颖的多模态生成评估框架，通过三角验证三个互补信号：（1）用于全局图像-文本分布对齐的互信息散度（Mutual Information Divergence，MID）；（2）基于DINO的循环生成图像感知相似性，用于视觉接地；（3）用于与人类参考进行上下文文本相似性比较的大语言模型（LLM）文本嵌入。对这些信号的校准融合使得RS能够提供更全面的评估。在Flickr8k基准上，RS实现了58.42的Kendall-τ系数，优于大多数先前方法，并且无需任务特定训练即可展示出与人类判断的卓越相关性。我们的框架通过同时彻底检查视觉准确性和文本质量，提供了更稳健和细致的评估，并在Conceptual Captions和MS COCO上表现一致。
英文摘要：
Evaluating image captions requires cohesive assessment of both visual semantics and language pragmatics, which is often not entirely captured by most metrics. As such metrics increasingly guide model development, benchmarking, and system optimization in multimodal AI, inaccuracies in evaluation can misrepresent true progress. We introduce Redemption Score(RS), a novel evaluation framework for multi-modal generation by triangulating three complementary signals: (1) Mutual Information Divergence (MID) for global image-text distributional alignment, (2) DINO-based perceptual similarity of cycle-generated images for visual grounding, and (3) LLM Text Embeddings for contextual text similarity against human references. A calibrated fusion of these signals allows RS to offer a more holistic assessment. On the Flickr8k benchmark, RS achieves a Kendall-$τ$ of 58.42, outperforming most prior methods and demonstrating superior correlation with human judgments without requiring task-specific training. Our framework provides a more robust and nuanced evaluation by thoroughly examining both the visual accuracy and text quality together, with consistent performance across Conceptual Captions and MS COCO.
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。
