# Improve Multi-Modal Embedding Learning via Explicit Hard Negative Gradient Amplifying

arXiv: https://arxiv.org/abs/2506.02020 · HTML: https://mlsanity.com/paper/2506.02020/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Placed in Document understanding and multimodal RAG, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 27 of its 29 references that are in the map.
- First posted on arXiv 28 May 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Document understanding and multimodal RAG](https://mlsanity.com/mllm/document-understanding-and-multimodal-rag/)

## Builds on (most cited first, 27 of 27)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [Reproducible scaling laws for contrastive language-image learning](https://mlsanity.com/paper/2212.07143/) (2022)
- [NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models](https://mlsanity.com/paper/2405.17428/) (2024)
- [VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documents](https://mlsanity.com/paper/2410.10594/) (2024)
- [VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Tasks](https://mlsanity.com/paper/2410.05160/) (2024)
- [Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup](https://mlsanity.com/paper/2101.06983/) (2021)
- [E5-V: Universal Embeddings with Multimodal Large Language Models](https://mlsanity.com/paper/2407.12580/) (2024)
- [Visual News: Benchmark and Challenges in News Image Captioning](https://mlsanity.com/paper/2010.03743/) (2020)
- [MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions](https://mlsanity.com/paper/2403.19651/) (2024)
- [EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters](https://mlsanity.com/paper/2402.04252/) (2024)
- [Breaking the Modality Barrier: Universal Embedding Learning with Multimodal LLMs](https://mlsanity.com/paper/2504.17432/) (2025)
- [VDocRAG: Retrieval-Augmented Generation over Visually-Rich Documents](https://mlsanity.com/paper/2504.09795/) (2025)
- [LLaVE: Large Language and Vision Embedding Models with Hardness-Weighted Contrastive Learning](https://mlsanity.com/paper/2503.04812/) (2025)
- [NoteLLM-2: Multimodal Large Representation Models for Recommendation](https://mlsanity.com/paper/2405.16789/) (2024)
- [NoteLLM: A Retrievable Large Language Model for Note Recommendation](https://mlsanity.com/paper/2403.01744/) (2024)
- [Breaking the Batch Barrier (B3) of Contrastive Learning via Smart Batch Mining](https://mlsanity.com/paper/2505.11293/) (2025)

## Built on by (most cited first, 23 of 23)

- [UniME-V2: MLLM-as-a-Judge for Universal Multimodal Embedding Learning](https://mlsanity.com/paper/2510.13515/) (2025)
- [Think Then Embed: Generative Context Improves Multimodal Embedding](https://mlsanity.com/paper/2510.05014/) (2025)
- [UME-R1: Exploring Reasoning-Driven Generative Multimodal Embeddings](https://mlsanity.com/paper/2511.00405/) (2025)
- [RzenEmbed: Towards Comprehensive Multimodal Retrieval](https://mlsanity.com/paper/2510.27350/) (2025)
- [Sculpting the Vector Space: Towards Efficient Multi-Vector Visual Document Retrieval via Prune-then-Merge Framework](https://mlsanity.com/paper/2602.19549/) (2026)
- [FreeRet: MLLMs as Training-Free Retrievers](https://mlsanity.com/paper/2509.24621/) (2025)
- [ReMatch: Boosting Representation through Matching for Multimodal Retrieval](https://mlsanity.com/paper/2511.19278/) (2025)
- [Unlocking Multimodal Document Intelligence: From Current Triumphs to Future Frontiers of Visual Document Retrieval](https://mlsanity.com/paper/2602.19961/) (2026)
- [TSEmbed: Unlocking Task Scaling in Universal Multimodal Embeddings](https://mlsanity.com/paper/2603.04772/) (2026)
- [FuseLIP: Multimodal Embeddings via Early Fusion of Discrete Tokens](https://mlsanity.com/paper/2506.03096/) (2025)
- [Magic-MM-Embedding: Towards Visual-Token-Efficient Universal Multimodal Embedding with MLLMs](https://mlsanity.com/paper/2602.05275/) (2026)
- [Adaptive Global and Fine-Grained Perceptual Fusion for MLLM Embeddings Compatible with Hard Negative Amplification](https://mlsanity.com/paper/2602.05729/) (2026)
- [Let Multimodal Embedders Learn When to Augment Query via Adaptive Query Augmentation](https://mlsanity.com/paper/2511.02358/) (2025)
- [PhotoBench: Beyond Visual Matching Towards Personalized Intent-Driven Photo Retrieval](https://mlsanity.com/paper/2603.01493/) (2026)
- [Enhancing Multimodal In-Context Learning via Inductive-Deductive Reasoning](https://mlsanity.com/paper/2605.02378/) (2026)
- [RAVEN: Long-Horizon Reasoning & Navigation with a Visuo-Spatio-Temporal Memory](https://mlsanity.com/paper/2606.25206/) (2026)
- [OSCAR: Optimization-Steered Agentic Planning for Composed Image Retrieval](https://mlsanity.com/paper/2602.08603/) (2026)
- [Illuminating Visual Identity in Universal Multimodal Embeddings](https://mlsanity.com/paper/2608.01794/) (2026)
- [Weaving Visual Narratives: Agentic Image Bundle Composition Beyond Atomic Visual Matching](https://mlsanity.com/paper/2608.28695/) (2026)
- [Evo-Retriever: LLM-Guided Curriculum Evolution with Viewpoint-Pathway Collaboration for Multimodal Document Retrieval](https://mlsanity.com/paper/2603.16455/) (2026)
- [Evaluation of Embedding-Based and Generative Methods for LLM-Driven Document Classification: Opportunities and Challenges](https://mlsanity.com/paper/2604.04997/) (2026)
- [AMBER: Multi-View Adaptive Budget Allocation for Listwise Vision-Language Reranking](https://mlsanity.com/paper/2610.02831/) (2026)
- [What Really Works for Universal Multimodal Embeddings? A Controlled Study of Contrastive Training Strategies](https://mlsanity.com/paper/2610.15035/) (2026)

## Abstract (arXiv)

With the rapid advancement of multi-modal large language models (MLLMs) in recent years, the foundational Contrastive Language-Image Pretraining (CLIP) framework has been successfully extended to MLLMs, enabling more powerful and universal multi-modal embeddings for a wide range of retrieval tasks. Despite these developments, the core contrastive learning paradigm remains largely unchanged from CLIP-style models to MLLMs. Within this framework, the effective mining of hard negative samples continues to be a critical factor for enhancing performance. Prior works have introduced both offline and online strategies for hard negative mining to improve the efficiency of contrastive learning. While these approaches have led to improved multi-modal embeddings, the specific contribution of each hard negative sample to the learning process has not been thoroughly investigated. In this work, we conduct a detailed analysis of the gradients of the info-NCE loss with respect to the query, positive, and negative samples, elucidating the role of hard negatives in updating model parameters. Building upon this analysis, we propose to explicitly amplify the gradients associated with hard negative samples, thereby encouraging the model to learn more discriminative embeddings. Our multi-modal embedding model, trained with the proposed Explicit Gradient Amplifier and based on the LLaVA-OneVision-7B architecture, achieves state-of-the-art performance on the MMEB benchmark compared to previous methods utilizing the same MLLM backbone. Furthermore, when integrated with our self-developed MLLM, QQMM, our approach attains the top rank on the MMEB leaderboard. Code and models are released on https://github.com/QQ-MM/QQMM-embed.