HOMIE: Histopathology Omni-modal Embedding For Pathology Composed Retrieval

Abstract

The integration of Artificial Intelligence (AI) into pathology faces a fundamental challenge: black-box predictive models lack transparency, while generative approaches risk clinical hallucination. A case-based retrieval paradigm offers a more interpretable alternative for clinical adoption. However, current SOTA models are constrained by dual-encoder architectures that cannot process the composed modality of real-world clinical queries. We formally define the task of Pathology Composed Retrieval (PCR). However, progress in this newly defined task is blocked by two critical challenges: (1) Multimodal Large Language Models (MLLMs) offer the necessary deep-fusion architecture but suffer from a critical Task Mismatch and Domain Mismatch. (2) No benchmark exists to evaluate such compositional queries. To solve these challenges, we propose HOMIE, a systematic framework that transforms a general MLLM into a specialized retrieval expert. HOMIE resolves the dual mismatch via a two-stage proces

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyzhou2025homie

Related papers

• Multimodal Learning For Scalable Representation Of High-dimensional Medical Data (2024)0.00
• Accurate And Scalable Multimodal Pathology Retrieval Via Attentive Vision-language Alignment (2025)2.26
• Generative Vector Search To Improve Pathology Foundation Models Across Multimodal Vision-language Tasks (2025)0.00
• Multimodal Whole Slide Foundation Model For Pathology (2024)12.99
• HMAR: Hierarchical Modality-aware Expert And Dynamic Routing Medical Image Retrieval Architecture (2026)0.00
• Pathalign: A Vision-language Model For Whole Slide Images In Histopathology (2024)0.00
• On The Importance Of Text Preprocessing For Multimodal Representation Learning And Pathology Report Generation (2025)0.00
• CREM: Compression-driven Representation Enhancement For Multimodal Retrieval And Comprehension (2026)0.00