XR: Cross-Modal Agents for Composed Image Retrieval

TL;DR

XR introduces a multi-agent, training-free framework for composed image retrieval that combines generative, matching, and reasoning agents to improve semantic and visual accuracy significantly.

Contribution

It proposes a novel multi-agent, training-free approach for composed image retrieval, enhancing semantic understanding and reasoning over existing embedding-based methods.

Findings

Achieves up to 38% improvement over baselines.

Each agent component is essential for performance.

Effective across multiple datasets: FashionIQ, CIRR, CIRCO.

Abstract

Retrieval is being redefined by agentic AI, demanding multimodal reasoning beyond conventional similarity-based paradigms. Composed Image Retrieval (CIR) exemplifies this shift as each query combines a reference image with textual modifications, requiring compositional understanding across modalities. While embedding-based CIR methods have achieved progress, they remain narrow in perspective, capturing limited cross-modal cues and lacking semantic reasoning. To address these limitations, we introduce XR, a training-free multi-agent framework that reframes retrieval as a progressively coordinated reasoning process. It orchestrates three specialized types of agents: imagination agents synthesize target representations through cross-modal generation, similarity agents perform coarse filtering via hybrid matching, and question agents verify factual consistency through targeted reasoning for fine filtering. Through progressive multi-agent coordination, XR iteratively refines retrieval to meet both semantic and visual query constraints, achieving up to a 38% gain over strong training-free and training-based baselines on FashionIQ, CIRR, and CIRCO, while ablations show each agent is essential. Code is available: https://01yzzyu.github.io/xr.github.io/.