Muco: Multi-turn Contrastive Learning For Multimodal Embedding Model
2026 · Geonmo Gu, Byeongho Heo, Jaemyung Yu, et al.
Abstract
Universal Multimodal embedding models built on Multimodal Large Language Models (MLLMs) have traditionally employed contrastive learning, which aligns representations of query-target pairs across different modalities. Yet, despite its empirical success, they are primarily built on a "single-turn" formulation where each query-target pair is treated as an independent data point. This paradigm leads to computational inefficiency when scaling, as it requires a separate forward pass for each pair and overlooks potential contextual relationships between multiple queries that can relate to the same context. In this work, we introduce Multi-Turn Contrastive Learning (MuCo), a dialogue-inspired framework that revisits this process. MuCo leverages the conversational nature of MLLMs to process multiple, related query-target pairs associated with a single image within a single forward pass. This allows us to extract a set of multiple query and target embeddings simultaneously, conditioned on a sha
Authors
(none)
Tags
Stats
Related papers
- Breaking The Modality Barrier: Universal Embedding Learning With Multimodal Llms (2025)4.52
- Unimoco: Unified Modality Completion For Robust Multi-modal Embeddings (2025)1.40
- Compressing Then Matching: An Efficient Pre-training Paradigm For Multimodal Embedding (2025)0.00
- Magic-mm-embedding: Towards Visual-token-efficient Universal Multimodal Embedding With Mllms (2026)0.00
- CREM: Compression-driven Representation Enhancement For Multimodal Retrieval And Comprehension (2026)0.00
- Mm-embed: Universal Multimodal Retrieval With Multimodal Llms (2024)0.00
- U-MARVEL: Unveiling Key Factors For Universal Multimodal Retrieval Via Embedding Learning With Mllms (2025)3.11
- Modality Curation: Building Universal Embeddings For Advanced Multimodal Information Retrieval (2025)0.00