Research map / Vision-language and multimodal / Human and embodied modeling
Mixture of Attribute-Aware Attention Experts for Fine-grained E-Commerce Composed Image Retrieval
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a mixture-of-attribute-aware-attention-experts model for fine-grained e-commerce composed image retrieval that exploits item text descriptions and handles multi-attribute confounding.
Problem
Composed image retrieval for e-commerce ignores readily available item textual descriptions, suffers from multi-attribute confounding (uncontrolled variation in unrelated attributes), and lacks explicit fine-grained attribute modeling.
Approach
Uses a mixture of attribute-aware attention experts so that each expert models a specific attribute, leveraging item descriptions to disentangle attributes in composed queries. Details are not available beyond the abstract.
Closest papers by meaning
- TIGER-FG: Text-Guided Implicit Fine-Grained Grounding for E-commerce Retrieval (NeurIPS 2026 · similarity 0.87)
- Transformer-Empowered Multi-Modal Item Embedding for Enhanced Image Search in E-commerce (AAAI 2024 · similarity 0.87)
- Instance-Level Composed Image Retrieval (NeurIPS 2025 · similarity 0.87)
- ConText-CIR: Learning from Concepts in Text for Composed Image Retrieval (CVPR 2025 · similarity 0.86)
- Data Roaming and Quality Assessment for Composed Image Retrieval (AAAI 2024 · similarity 0.85)
- ENCODER: Entity Mining and Modification Relation Binding for Composed Image Retrieval (AAAI 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.