MELT: Improve Composed Image Retrieval Via The Modification Frequentation-rarity Balance Network

Abstract

Composed Image Retrieval (CIR) uses a reference image and a modification text as a query to retrieve a target image satisfying the requirement of ``modifying the reference image according to the text instructions''. However, existing CIR methods face two limitations: (1) frequency bias leading to ``Rare Sample Neglect'', and (2) susceptibility of similarity scores to interference from hard negative samples and noise. To address these limitations, we confront two key challenges: asymmetric rare semantic localization and robust similarity estimation under hard negative samples. To solve these challenges, we propose the Modification frEquentation-rarity baLance neTwork MELT. MELT assigns increased attention to rare modification semantics in multimodal contexts while applying diffusion-based denoising to hard negative samples with high similarity scores, enhancing multimodal fusion and matching. Extensive experiments on two CIR benchmarks validate the superior performance of MELT. Codes ar

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyqiu2026melt

Related papers

• HINT: Composed Image Retrieval With Dual-path Compositional Contextualized Network (2026)0.78
• DAFM: Dynamic Adaptive Fusion For Multi-model Collaboration In Composed Image Retrieval (2025)0.00
• Far-net: Multi-stage Fusion Network With Enhanced Semantic Alignment And Adaptive Reconciliation For Composed Image Retrieval (2025)0.00
• Finecir: Explicit Parsing Of Fine-grained Modification Semantics For Composed Image Retrieval (2025)2.16
• NCL-CIR: Noise-aware Contrastive Learning For Composed Image Retrieval (2025)2.26
• TMCIR: Token Merge Benefits Composed Image Retrieval (2025)0.00
• HABIT: Chrono-synergia Robust Progressive Learning Framework For Composed Image Retrieval (2026)2.35
• INTENT: Invariance And Discrimination-aware Noise Mitigation For Robust Composed Image Retrieval (2026)0.00