Papers › Named Entity and Relation Extraction with Multi-Modal Retrieval

Named Entity and Relation Extraction with Multi-Modal Retrieval

3 Dec 2022arXiv:2212.01612archive 2025-07-28

Xinyu Wang, Jiong Cai, Yong Jiang, Pengjun Xie, Kewei Tu, Wei Lu

Multi-modal named entity recognition (NER) and relation extraction (RE) aim to leverage relevant image information to improve the performance of NER and RE. Most existing efforts largely focused on directly extracting potentially useful information from images (such as pixel-level features, identified objects, and associated captions). However, such extraction processes may not be knowledge aware, resulting in information that may not be highly relevant. In this paper, we propose a novel Multi-modal Retrieval based framework (MoRe). MoRe contains a text retrieval module and an image-based retrieval module, which retrieve related knowledge of the input text and image in the knowledge corpus respectively. Next, the retrieval results are sent to the textual and visual models respectively for predictions. Finally, a Mixture of Experts (MoE) module combines the predictions from the two models to make the final decision. Our experiments show that both our textual model and visual model can achieve state-of-the-art performance on four multi-modal NER datasets and one multi-modal RE dataset. With MoE, the model performance can be further improved and our analysis demonstrates the benefits of integrating both textual and visual cues for such tasks.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

modelscope/adaseq officialmentioned in paperpytorchApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Mixture-of-ExpertsMulti-modal Named Entity RecognitionNamed Entity RecognitionNamed Entity Recognition (NER)Relation ExtractionRetrievalText Retrieval

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Multi-modal Named Entity Recognition SNAP (MNER) MoRe-MoE F1 91.10 #1 of 6 Archive leaderboard report
Multi-modal Named Entity Recognition SNAP (MNER) MoRe-Image F1 90.20 #2 of 6 Archive leaderboard report
Multi-modal Named Entity Recognition SNAP (MNER) MoRe-Text F1 90.09 #4 of 6 Archive leaderboard report
Multi-modal Named Entity Recognition SNAP (MNER) BERT-CRF F1 89.65 #5 of 6 Archive leaderboard report
Multi-modal Named Entity Recognition Twitter-15 MoRe-MoE F1 79.21 #2 of 6 Archive leaderboard report
Multi-modal Named Entity Recognition Twitter-15 MoRe-Image F1 78.13 #3 of 6 Archive leaderboard report
Multi-modal Named Entity Recognition Twitter-15 MoRe-Text F1 77.91 #4 of 6 Archive leaderboard report
Multi-modal Named Entity Recognition Twitter-15 BERT-CRF F1 77.04 #5 of 6 Archive leaderboard report
Multi-modal Named Entity Recognition Twitter-17 MoRe-MoE F1 90.67 #2 of 2 Archive leaderboard report
Multi-modal Named Entity Recognition Twitter-2017 MoRe-MoE F1 90.67 #2 of 3 Archive leaderboard report
Multi-modal Named Entity Recognition WikiDiverse MoRe-MoE F1 79.33 #1 of 5 Archive leaderboard report
Multi-modal Named Entity Recognition WikiDiverse MoRe-Text F1 77.97 #2 of 5 Archive leaderboard report
Multi-modal Named Entity Recognition WikiDiverse MoRe-Image F1 77.46 #3 of 5 Archive leaderboard report
Multi-modal Named Entity Recognition WikiDiverse BERT-CRF F1 76.58 #5 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections