Reconstructing Content via Collaborative Attention to Improve Multimodal Embedding Quality

• URL: http://arxiv.org/abs/2603.01471v1
• Date: Mon, 02 Mar 2026 05:34:45 GMT
• Title: Reconstructing Content via Collaborative Attention to Improve Multimodal Embedding Quality
• Authors: Jiahan Chen, Da Li, Hengran Zhang, Yinqiong Cai, Lixin Su, Jiafeng Guo, Daiting Shi, Dawei Yin, Keping Bi,
• Abstract summary: CoCoA is a Content reconstruction pre-training paradigm based on Collaborative Attention for multimodal embedding optimization.<n>We introduce an EOS-based reconstruction task, encouraging the model to reconstruct input from the corresponding EOS> embeddings.<n>Experiments on MMEB-V1 demonstrate that CoCoA built upon Qwen2-VL and Qwen2.5-VL significantly improves embedding quality.
• Score: 59.651410243721045
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multimodal embedding models, rooted in multimodal large language models (MLLMs), have yielded significant performance improvements across diverse tasks such as retrieval and classification. However, most existing approaches rely heavily on large-scale contrastive learning, with limited exploration of how the architectural and training paradigms of MLLMs affect embedding quality. While effective for generation, the causal attention and next-token prediction paradigm of MLLMs does not explicitly encourage the formation of globally compact representations, limiting their effectiveness as multimodal embedding backbones. To address this, we propose CoCoA, a Content reconstruction pre-training paradigm based on Collaborative Attention for multimodal embedding optimization. Specifically, we restructure the attention flow and introduce an EOS-based reconstruction task, encouraging the model to reconstruct input from the corresponding <EOS> embeddings. This drives the multimodal model to compress the semantic information of the input into the <EOS> token, laying the foundations for subsequent contrastive learning. Extensive experiments on MMEB-V1 demonstrate that CoCoA built upon Qwen2-VL and Qwen2.5-VL significantly improves embedding quality. Results validate that content reconstruction serves as an effective strategy to maximize the value of existing data, enabling multimodal embedding models generate compact and informative representations, raising their performance ceiling.

Related papers

• Magic-MM-Embedding: Towards Visual-Token-Efficient Universal Multimodal Embedding with MLLMs [10.443777669301983]
  Multimodal Large Language Models (MLLMs) have shown immense promise in universal multimodal retrieval.<n>But their practical application is often hindered by the substantial computational cost incurred from processing a large number of tokens from visual inputs.<n>We propose Magic-MM-Embedding, a series of novel models that achieve both high efficiency and state-of-the-art performance in universal multimodal embedding.
  arXiv  Detail & Related papers  (2026-02-05T04:01:01Z)
• ParaUni: Enhance Generation in Unified Multimodal Model with Reinforcement-driven Hierarchical Parallel Information Interaction [55.21514454560188]
  Unified multimodal models significantly improve visual generation by combining vision-grained models (VLMs) with diffusion models.<n>Existing methods struggle to fully balance sufficient interaction and flexible implementation due to vast representation difference.<n>We propose textbfParaUni, which extracts features from variants VLM's layers in a textbfParallel way for comprehensive information interaction.
  arXiv  Detail & Related papers  (2025-12-05T04:41:57Z)
• Multi-Aspect Cross-modal Quantization for Generative Recommendation [27.92632297542123]
  We propose Multi-Aspect Cross-modal quantization for generative Recommendation (MACRec)<n>We first introduce cross-modal quantization during the ID learning process, which effectively reduces conflict rates.<n>We also incorporate multi-aspect cross-modal alignments, including the implicit and explicit alignments.
  arXiv  Detail & Related papers  (2025-11-19T04:55:14Z)
• Compression then Matching: An Efficient Pre-training Paradigm for Multimodal Embedding [53.18433310890516]
  Vision-language models advance multimodal representation learning by acquiring transferable semantic embeddings.<n>We propose CoMa, a compressed pre-training phase, which serves as a warm-up stage for contrastive learning.
  arXiv  Detail & Related papers  (2025-11-11T17:23:02Z)
• MoCa: Modality-aware Continual Pre-training Makes Better Bidirectional Multimodal Embeddings [75.0617088717528]
  MoCa is a framework for transforming pre-trained VLM backbones into effective bidirectional embedding models.<n>MoCa consistently improves performance across MMEB and ViDoRe-v2 benchmarks, achieving new state-of-the-art results.
  arXiv  Detail & Related papers  (2025-06-29T06:41:00Z)
• Will Pre-Training Ever End? A First Step Toward Next-Generation Foundation MLLMs via Self-Improving Systematic Cognition [89.50068130832635]
  Self-Improving cognition (SIcog) is a self-learning framework for constructing next-generation foundation MLLMs by multimodal knowledge.<n>We propose Chain-of-Description for step-by-step visual understanding and integrate structured Chain-of-Thought (CoT) reasoning to support in-depth multimodal reasoning.<n>Experiments demonstrate SIcog's effectiveness in developing MLLMs with enhanced multimodal cognition.
  arXiv  Detail & Related papers  (2025-03-16T00:25:13Z)
• RA-BLIP: Multimodal Adaptive Retrieval-Augmented Bootstrapping Language-Image Pre-training [55.54020926284334]
  Multimodal Large Language Models (MLLMs) have recently received substantial interest, which shows their emerging potential as general-purpose models for various vision-language tasks. Retrieval augmentation techniques have proven to be effective plugins for both LLMs and MLLMs. In this study, we propose multimodal adaptive Retrieval-Augmented Bootstrapping Language-Image Pre-training (RA-BLIP), a novel retrieval-augmented framework for various MLLMs.
  arXiv  Detail & Related papers  (2024-10-18T03:45:19Z)
• MMEvol: Empowering Multimodal Large Language Models with Evol-Instruct [148.39859547619156]
  We propose MMEvol, a novel multimodal instruction data evolution framework.<n>MMEvol iteratively improves data quality through a refined combination of fine-grained perception, cognitive reasoning, and interaction evolution.<n>Our approach reaches state-of-the-art (SOTA) performance in nine tasks using significantly less data compared to state-of-the-art models.
  arXiv  Detail & Related papers  (2024-09-09T17:44:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.