Beyond Global Similarity: Towards Fine-Grained, Multi-Condition Multimodal Retrieval

TL;DR

MCMR (Multi-Conditional Multimodal Retrieval): a large-scale benchmark designed to evaluate fine-grained, multi-condition cross-modal retrieval under natural-language queries and establishes a challenging and diagnostic benchmark for advancing multimodal retrieval toward compositional, constraint-aware, and interpretable understanding.

Abstract

Recent advances in multimodal large language models (MLLMs) have substantially expanded the capabilities of multimodal retrieval, enabling systems to align and retrieve information across visual and textual modalities. Yet, existing benchmarks largely focus on coarse-grained or single-condition alignment, overlooking real-world scenarios where user queries specify multiple interdependent constraints across modalities. To bridge this gap, we introduce MCMR (Multi-Conditional Multimodal Retrieval): a large-scale benchmark designed to evaluate fine-grained, multi-condition cross-modal retrieval under natural-language queries. MCMR spans five product domains: upper and bottom clothing, jewelry, shoes, and furniture. It also preserves rich long-form metadata essential for compositional matching. Each query integrates complementary visual and textual attributes, requiring models to jointly satisfy all specified conditions for relevance. We benchmark a diverse suite of MLLM-based multimodal retrievers and vision-language rerankers to assess their condition-aware reasoning abilities. Experimental results reveal: (i) distinct modality asymmetries across models; (ii) visual cues dominate early-rank precision, while textual metadata stabilizes long-tail ordering; and (iii) MLLM-based pointwise rerankers markedly improve fine-grained matching by explicitly verifying query-candidate consistency. Overall, MCMR establishes a challenging and diagnostic benchmark for advancing multimodal retrieval toward compositional, constraint-aware, and interpretable understanding. Our code and dataset is available at https://github.com/EIT-NLP/MCMR

Figures (10)

• Figure 1: Multi-condition multimodal retrieval driven by natural-language queries with fine-grained visual and textual constraints.
• Figure 2: Construction pipeline of MCMR. Visual and textual features are extracted by Qwen-VL and Qwen-Instruct, validated through two judging stages, and composed into queries for fine-grained retrieval evaluation.
• Figure 3: Retrieval performance (Recall@10) under different query regimes. The complete results are provided in the supplementary material.
• Figure 4: Recall@10 under varying numbers of compositional constraints ($k_T = k_I \in {1,2,3,4,5}$). Candidates are fixed with fused image–text metadata.
• Figure 5: Prompt for Image Attribute Extraction used in image-side annotation.