natural-language-processing

Multi-modal Entity Alignment

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in multi-modal-entity-alignment

Trend

Dataset

Best Model

Actions

UMVM-dbp-zh-en

UMVM-dbp-ja-en

UMVM-dbp-fr-en

Libraries

i

Use these libraries to find multi-modal-entity-alignment models and implementations

Datasets

Subtasks

No subtasks available.

Most implemented papers

Visual Pivoting for (Unsupervised) Entity Alignment

Dan Roth, Nigel Collier, Muhao Chen, Fangyu Liu•Sun Sep 27 2020

This work shows that images are particularly useful to align long-tail KG entities, which inherently lack the structural contexts necessary for capturing the correspondences, and provides EVA, a completely unsupervised solution to this problem.

152

Content

UMVM-dbp-fr-en

UMVM-oea-en-fr

UMVM-oea-en-de

UMVM-oea-d-w-v1

UMVM-oea-d-w-v2

0

Paper Graph

MMEA: Entity Alignment for Multi-modal Knowledge Graph

Enhong Chen, Liyi Chen, Zhi Li, Yijun Wang, Tong Xu, Zhefeng Wang•Thu Aug 27 2020

This paper proposes a novel solution called Multi-Modal Entity Alignment (MMEA) to address the problem of entity alignment in a multi-modal view and designs a novel multi- modal knowledge embedding method to generate the entity representations of relational, visual and numerical knowledge.

91 0

Paper Graph

Rethinking Uncertainly Missing and Ambiguous Visual Modality in Multi-Modal Entity Alignment

Hua-zeng Chen, Y. Li, Zhuo Chen, Lingbing Guo, Yin Fang, Yichi Zhang, Jiaoyan Chen, Jeff Z. Pan, Wen Zhang•Sat Jul 29 2023

UMAEA is introduced, a robust multi-modal entity alignment approach designed to tackle uncertainly missing and ambiguous visual modalities, which consistently achieves SOTA performance across all 97 benchmark splits, significantly surpassing existing baselines with limited parameters and time consumption, while effectively alleviating the identified limitations of other models.

47 0

Paper Graph

Multi-modal Contrastive Representation Learning for Entity Alignment

Ziheng Zhang, Yefeng Zheng, Zhenxi Lin, Meng Wang, Yinghui Shi, Xianlong Wu•Thu Sep 01 2022

Different from previous works, MCLEA considers task-oriented modality and models the inter-modal relationships for each entity representation, and outperforms state-of-the-art baselines on public datasets under both supervised and unsupervised settings.

75 0

Paper Graph

Multi-Modal Knowledge Graph Transformer Framework for Multi-Modal Entity Alignment

Lihong Wang, Qian Li, Cheng Ji, Shu Guo, Zhaoji Liang, Jianxin Li•Mon Oct 09 2023

A novel MMEA transformer is proposed, called MoAlign, that hierarchically introduces neighbor features, multi-modal attributes, and entity types to enhance the alignment task and outperforms strong competitors and achieves excellent entity alignment performance.

20 0

Paper Graph

Towards Semantic Consistency: Dirichlet Energy Driven Robust Multi-Modal Entity Alignment

Qi Qi, Haifeng Sun, Jianxin Liao, Jingyu Wang, Yuanyi Wang, Jiabo Wang, Wei Tang, Shaoling Sun•Tue Jan 30 2024

DESAlign is proposed, a robust method addressing the over-smoothing caused by semantic inconsistency and interpolating missing semantics using existing modalities using existing modalities, and a training strategy for multi-modal knowledge graph learning based on the proposed generalizable theoretical principle.

21 0

Paper Graph

The Power of Noise: Toward a Unified Multi-modal Knowledge Graph Representation Framework

Wen Zhang, Hua-zeng Chen, Zhuo Chen, Lingbing Guo, Yin Fang, Yichi Zhang, Jiaoyan Chen•Sun Dec 31 2023

This work proposes a novel SnAg method that utilizes a Transformer-based architecture equipped with modality-level noise masking for the robust integration of multi-modal entity features in KGs, and achieves SOTA performance across a total of ten datasets.

4 0

Paper Graph

Universal Multi-modal Entity Alignment via Iteratively Fusing Modality Similarity Paths

Xiaoze Liu, Zhuo Chen, Lingbing Guo, Tao Gui, Qi Zhang, Bolin Zhu, Xin Mao•Sun Oct 08 2023

PathFusion is proposed, consisting of two main components: MSP, a unified modeling approach that simplifies the alignment process by constructing paths connecting entities and modality nodes to represent multiple modalities; IRF, an iterative fusion method that effectively combines information from different modalities using the path as an information carrier.

2 0

Paper Graph

MEAformer: Multi-modal Entity Alignment Transformer for Meta Modality Hybrid

Hua-zeng Chen, Zhuo Chen, Lingbing Guo, Yin Fang, Jiaoyan Chen, Jeff Z. Pan, Wen Zhang, Yuxia Geng, Yichi Zhang, Yufen Huang, Wenting Song•Wed Dec 28 2022

MEAformer is introduced, a mlti-modal entity alignment transformer approach for meta modality hybrid, which dynamically predicts the mutual correlation coefficients among modalities for more fine-grained entity-level modality fusion and alignment.

76 0

Paper Graph

Multi-modal Siamese Network for Entity Alignment

Liyi Chen, Zhi Li, N. Yuan, Zhefeng Wang, Tong Xu, Han Wu, Enhong Chen•Sat Aug 13 2022

A novel Multi- modal Siamese Network for Entity Alignment (MSNEA) is proposed to align entities in different MMKGs, in which multi-modal knowledge could be comprehensively leveraged by the exploitation of inter-modAL effect.

101 0

Paper Graph

Adding a benchmark result helps the community track progress.