paper-with-me

홈 › Papers

UniNote: A Unified Embedding Model for Multimodal Representation and Ranking

2026-05-28 · Jinghan Zhao, Wenwei Jin, Anqi Li, Jintao Tong, Luya Mo, Jiawei Li, Bin Li, Yao Hu arxiv

Item-to-Item (I2I) retrieval is a fundamental part of modern content platforms, supporting critical industrial workflows from recommendation engines to content auditing. While multimodal embedding methods have advanced general retrieval, they often falter in I2I scenarios due to the challenges of balancing global content representation with fine-grained local retrieval, the systemic inefficiency of decoupled embedding-and-ranking pipelines, and the inherent trade-offs between model precision and serving latency. To solve these issues, we propose \textbf{UniNote}, a unified embedding model designed for industrial I2I retrieval. Tailored retrieval strategies are introduced to support representation learning over complex, multimodal content at varying granularities. To operationalize these strategies, UniNote employs a two-stage training paradigm: the first stage leverages contrastive SFT to establish robust base embeddings, while the second stage refines ranking quality through a reinforcement learning (RL) process that aligns the model with content relevance. Our results show that UniNote achieves SOTA performance across diverse I2I tasks. Deployed at Xiaohongshu and integrated with Matryoshka Representation Learning (MRL), UniNote achieved significant improvements in retrieval quality and cost efficiency in large-scale applications.

📄 PDF Abstract BibTeX arXiv:2605.29287

Code (0)

등록된 구현이 없습니다.

Tasks

Representation LearningReinforcement Learning

Similar Papers 제목 키워드 기반

Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Ranking

2026-01-08 · Mingxin Li, Yanzhao Zhang, Dingkun Long, Keqin Chen 외 arxiv

In this report, we introduce the Qwen3-VL-Embedding and Qwen3-VL-Reranker model series, the latest extensions of the Qwen family built on the Qwen3-VL foundation model. Together, they provide an end-to-end pipeline for h…

Visual Question AnsweringRepresentation LearningText Retrieval

UMER: Unifying Embedding and Ranking via Pair-Aware Discriminative Reasoning for Universal Multimodal Retrieval

2026-08-19 · Libiao Chen, Xiyang Liu, Yanheng Wei, Tao Wang 외 arxiv

Universal multimodal retrieval aims to support diverse instruction-aware retrieval tasks, demanding both efficient corpus-scale matching and fine-grained semantic reasoning. Recent MLLM-based embedding methods typically …

Incorporating Dense Knowledge Alignment into Unified Multimodal Representation Models

2025-01-01 · CVPR 2025 1 · Yuhao Cui, Xinxing Zu, Wenhua Zhang, Zhongzhou Zhao 외

Leveraging Large Language Models (LLMs) for text representation has achieved significant success, but the exploration of using Multimodal LLMs (MLLMs) for multimodal representation remains limited. Previous MLLM-base…

Contrastive LearningCross-Modal RetrievalRetrieval

PixelBytes: Catching Unified Embedding for Multimodal Generation

2024-09-03 · Fabien Furfaro

This report introduces PixelBytes Embedding, a novel approach for unified multimodal representation learning. Our method captures diverse inputs in a single, cohesive representation, enabling emergent properties for mult…

Mambamultimodal generationRepresentation LearningState Space Models

Unified Interactive Multimodal Moment Retrieval via Cascaded Embedding-Reranking and Temporal-Aware Score Fusion

2025-12-15 · Toan Le Ngo Thanh, Phat Ha Huu, Tan Nguyen Dang Duy, Thong Nguyen Le Minh 외 arxiv

The exponential growth of video content has created an urgent need for efficient multimodal moment retrieval systems. However, existing approaches face three critical challenges: (1) fixed-weight fusion strategies fail a…

Moment Retrieval