paper-with-me

Papers

Magic-MM-Embedding: Towards Visual-Token-Efficient Universal Multimodal Embedding with MLLMs

2026-02-05 · Qi Li, Yanzhe Zhao, Yongxin Zhou, Yameng Wang, Yandong Yang, Yuanjia Zhou, Jinxiang Liu arxiv

Multimodal Large Language Models (MLLMs) have shown immense promise in universal multimodal retrieval, which aims to find relevant items of various modalities for a given query. However, their practical application is often hindered by the substantial computational cost incurred from processing a large number of tokens from visual inputs. In this paper, 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. Our approach is built on two synergistic pillars: (1) a highly efficient MLLM architecture incorporating visual token compression to drastically reduce inference latency and training time, and (2) a multi-stage progressive training strategy designed to not only recover but significantly boost performance. This coarse-to-fine training paradigm begins with extensive continued training to restore multimodal understanding and generation capabilities, progresses to large-scale contrastive pretraining and hard negative mining to enhance discriminative power, and culminates in a task-aware fine-tuning stage guided by an MLLM-as-a-Judge for precise data curation. Comprehensive experiments show that our model outperforms existing methods by a large margin while being more inference-efficient.

📄 PDF Abstract BibTeX arXiv:2602.05275

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

MAGIC: Multimodal Alignment & Grounding-aware Instruction Coreset for Vision-Language Models

2026-05-25 · Shristi Das Biswas, Kaushik Roy arxiv

Instruction tuning of large vision-language models (LVLMs) increasingly depends on massive multimodal corpora, yet these datasets contain samples with substantial redundancy, low visual dependency, and highly imbalanced …

Multimodal Reasoning

Jailbreaking LLMs' Safeguard with Universal Magic Words for Text Embedding Models

2025-01-30 · Haoyu Liang, Youran Sun, Yunfeng Cai, Jun Zhu 외

The security issue of large language models (LLMs) has gained wide attention recently, with various defense mechanisms developed to prevent harmful output, among which safeguards based on text embedding models serve as a…

MagicVL-2B: Empowering Vision-Language Models on Mobile Devices with Lightweight Visual Encoders via Curriculum Learning

2025-08-03 · Yi Liu, Xiao Xu, Zeyu Xu, Meng Zhang 외 arxiv

Vision-Language Models (VLMs) have achieved remarkable breakthroughs in recent years, enabling a diverse array of applications in everyday life. However, the substantial computational and storage demands of VLMs pose sig…

TTE-Flash: Accelerating Reasoning-based Multimodal Representations via Think-Then-Embed Tokens

2026-05-15 · Jianpeng Cheng, Xian Wu, Jiangfan Zhang, Wentao Bao 외 arxiv

Recent research has demonstrated that Universal Multimodal Embedding (UME) benefits significantly from Chain-of-Thought (CoT) reasoning. In this paradigm, a generative model produces explicit reasoning traces for a multi…

Bridging Modalities, Spanning Time: Structured Memory for Ultra-Long Agentic Video Reasoning

2026-05-08 · Jiazheng Li, Chi-Hao Wu, Yunze Liu, Kaize Ding 외 arxiv

Understanding ultra-long videos such as egocentric recordings, live streams, or surveillance footage spanning days to weeks, remains a challenge. For current multimodal LLMs: even with million-token context windows, fram…

Cross-Modal Retrieval