MMEL: A Joint Learning Framework for Multi-Mention Entity Linking

Chengmei Yang, Bowei He, Yimeng Wu, Chao Xing, Lianghua He, Chen Ma
Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence, PMLR 216:2411-2421, 2023.

Abstract

Entity linking, bridging mentions in the contexts with their corresponding entities in the knowledge bases, has attracted wide attention due to many potential applications. Recently, plenty of multimodal entity linking approaches have been proposed to take full advantage of the visual information rather than solely the textual modality. Although feasible, these methods mainly focus on the single-mention scenarios and neglect the scenarios where multiple mentions exist simultaneously in the same context, which limits the performance. In fact, such multi-mention scenarios are pretty common in public datasets and real-world applications. To solve this challenge, we first propose a joint feature extraction module to learn the representations of context and entity candidates, from both the visual and textual perspectives. Then, we design a pairwise training scheme (for training) and a multi-mention collaborative ranking method (for testing) to model the potential connections between different mentions. We evaluate our method on a public dataset and a self-constructed dataset, NYTimes-MEL, under both text-only and multimodal scenarios. The experimental results demonstrate that our method can largely outperform the state-of-the-art methods, especially in multi-mention scenarios. Our dataset and source code are publicly available at https://github.com/ycm094/MMEL-main.

Cite this Paper

BibTeX
@InProceedings{pmlr-v216-yang23d, title = {{MMEL}: A Joint Learning Framework for Multi-Mention Entity Linking}, author = {Yang, Chengmei and He, Bowei and Wu, Yimeng and Xing, Chao and He, Lianghua and Ma, Chen}, booktitle = {Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence}, pages = {2411--2421}, year = {2023}, editor = {Evans, Robin J. and Shpitser, Ilya}, volume = {216}, series = {Proceedings of Machine Learning Research}, month = {31 Jul--04 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v216/yang23d/yang23d.pdf}, url = {https://proceedings.mlr.press/v216/yang23d.html}, abstract = {Entity linking, bridging mentions in the contexts with their corresponding entities in the knowledge bases, has attracted wide attention due to many potential applications. Recently, plenty of multimodal entity linking approaches have been proposed to take full advantage of the visual information rather than solely the textual modality. Although feasible, these methods mainly focus on the single-mention scenarios and neglect the scenarios where multiple mentions exist simultaneously in the same context, which limits the performance. In fact, such multi-mention scenarios are pretty common in public datasets and real-world applications. To solve this challenge, we first propose a joint feature extraction module to learn the representations of context and entity candidates, from both the visual and textual perspectives. Then, we design a pairwise training scheme (for training) and a multi-mention collaborative ranking method (for testing) to model the potential connections between different mentions. We evaluate our method on a public dataset and a self-constructed dataset, NYTimes-MEL, under both text-only and multimodal scenarios. The experimental results demonstrate that our method can largely outperform the state-of-the-art methods, especially in multi-mention scenarios. Our dataset and source code are publicly available at https://github.com/ycm094/MMEL-main.} }
Endnote
%0 Conference Paper %T MMEL: A Joint Learning Framework for Multi-Mention Entity Linking %A Chengmei Yang %A Bowei He %A Yimeng Wu %A Chao Xing %A Lianghua He %A Chen Ma %B Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2023 %E Robin J. Evans %E Ilya Shpitser %F pmlr-v216-yang23d %I PMLR %P 2411--2421 %U https://proceedings.mlr.press/v216/yang23d.html %V 216 %X Entity linking, bridging mentions in the contexts with their corresponding entities in the knowledge bases, has attracted wide attention due to many potential applications. Recently, plenty of multimodal entity linking approaches have been proposed to take full advantage of the visual information rather than solely the textual modality. Although feasible, these methods mainly focus on the single-mention scenarios and neglect the scenarios where multiple mentions exist simultaneously in the same context, which limits the performance. In fact, such multi-mention scenarios are pretty common in public datasets and real-world applications. To solve this challenge, we first propose a joint feature extraction module to learn the representations of context and entity candidates, from both the visual and textual perspectives. Then, we design a pairwise training scheme (for training) and a multi-mention collaborative ranking method (for testing) to model the potential connections between different mentions. We evaluate our method on a public dataset and a self-constructed dataset, NYTimes-MEL, under both text-only and multimodal scenarios. The experimental results demonstrate that our method can largely outperform the state-of-the-art methods, especially in multi-mention scenarios. Our dataset and source code are publicly available at https://github.com/ycm094/MMEL-main.
APA
Yang, C., He, B., Wu, Y., Xing, C., He, L. & Ma, C.. (2023). MMEL: A Joint Learning Framework for Multi-Mention Entity Linking. Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 216:2411-2421 Available from https://proceedings.mlr.press/v216/yang23d.html.

Related Material