发表机构
School of Computer Science and Technology, Soochow University; INSTITUT NATIONAL DES SCIENCES APPLIQUEES DE LYON（苏州大学计算机科学与技术学院; 里昂国立应用科学学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出无需训练的AutoConcept重排序器，用于元数据可用的组合图像检索，通过过滤噪声概念、激活正约束等方式提升检索效果，在FashionIQ等数据集上取得显著增益，验证了概念级重排序的有效性。
AI 中文摘要
组合图像检索（CIR）从参考图像和文本修改中检索目标图像。本文研究元数据可用的CIR重排序：先由固定CIR模型返回候选池，再利用图库元数据进行第二阶段的概念引导评分。我们提出AutoConcept，这是一种无需训练的重排序器，可将概念证据转换为可解释的内存。AutoConcept过滤噪声概念，通过辅助负惩罚激活查询相关的正约束，并通过推理时校准将基础检索分数与基于元数据的概念-候选对齐相结合。在FashionIQ数据集上，AutoConcept相比WeiMoCIR在早期排名上取得显著提升，在LinCIR候选池上实现一致的插件增益。元数据感知控制实验表明，结构化概念内存能提供超出直接查询文本和提取属性匹配的信号，仅查询的变体进一步验证了概念级重排序的有效性。补充的真实人类概念标签研究显示，该内存接口可使用参与者提供的证据。这些结果表明AutoConcept是适用于具有可用元数据的产品风格CIR图库的可解释概念内存重排序器。
英文摘要
Composed image retrieval (CIR) retrieves a target image from a reference image and a text modification. This paper studies metadata-available CIR reranking, where a fixed CIR model first returns a candidate pool and gallery metadata is then used for second-stage concept-guided scoring. We introduce AutoConcept, a training-free reranker that converts concept evidence into an interpretable memory. AutoConcept filters noisy concepts, activates query-relevant positive constraints with an auxiliary negative penalty, and combines base retrieval scores with metadata-based concept-candidate alignment through inference-time calibration. On FashionIQ, AutoConcept yields significant early-rank improvements over WeiMoCIR and consistent plug-in gains on LinCIR candidate pools. Metadata-aware controls show that structured concept memory adds signal beyond direct query-text and extracted-attribute matching, while a query-only variant further supports the effectiveness of concept-level reranking. A supplementary real-human concept-label study indicates that the same memory interface can consume participant-provided evidence. These results position AutoConcept as an interpretable concept-memory reranker for product-style CIR galleries with available metadata.
CommentsAccepted regular paper at PRICAI 2026. 16 pages, 4 figures
