Geo-Encoder: A Chunk-Argument Bi-Encoder Framework for Chinese Geographic Re-Ranking
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
Geo-Encoder : A Chunk-Argument Bi-Encoder Framework for Chinese Geographic Re-Ranking. / Cao, Yong ; Ding, Ruixue ; Chen, Boli ; Li, Xianzhi; Chen, Min; Hershcovich, Daniel; Xie, Pengjun ; Huang, Fei .
Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics (ACL), 2024. p. 1516–1530.Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
Association for Computational Linguistics - EACL 2024, St. Julian’s, Malta, 17/03/2024. <https://aclanthology.org/2024.eacl-long.91/>
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Geo-Encoder
T2 - 18th Conference of the European Chapter of the<br/>Association for Computational Linguistics - EACL 2024
AU - Cao, Yong
AU - Ding, Ruixue
AU - Chen, Boli
AU - Li, Xianzhi
AU - Chen, Min
AU - Hershcovich, Daniel
AU - Xie, Pengjun
AU - Huang, Fei
PY - 2024
Y1 - 2024
N2 - Chinese geographic re-ranking task aims to find the most relevant addresses among retrieved candidates, which is crucial for location-related services such as navigation maps. Unlike the general sentences, Chinese geographic contexts are closely intertwined with geographical concepts, from general spans (e.g., province) to specific spans (e.g., road). Given this feature, we propose an innovative framework, namely Geo-Encoder, to more effectively integrate Chinese geographical semantics into re-ranking pipelines. Our methodology begins by employing off-the-shelf tools to associate text with geographical spans, treating them as chunking units. Then, we present a multi-task learning module to simultaneously acquire an effective attention matrix that determines chunk contributions to geographic representations. Furthermore, we put forth an asynchronous update mechanism for the proposed task, aiming to guide the model to focus on specific chunks. Experiments on two Chinese benchmark datasets, show that the Geo-Encoder achieves significant improvements when compared to state-of-the-art baselines. Notably, it leads to a substantial improvement in the Hit@1 score of MGEO-BERT, increasing it by 6.22% from 62.76 to 68.98 on the GeoTES dataset.
AB - Chinese geographic re-ranking task aims to find the most relevant addresses among retrieved candidates, which is crucial for location-related services such as navigation maps. Unlike the general sentences, Chinese geographic contexts are closely intertwined with geographical concepts, from general spans (e.g., province) to specific spans (e.g., road). Given this feature, we propose an innovative framework, namely Geo-Encoder, to more effectively integrate Chinese geographical semantics into re-ranking pipelines. Our methodology begins by employing off-the-shelf tools to associate text with geographical spans, treating them as chunking units. Then, we present a multi-task learning module to simultaneously acquire an effective attention matrix that determines chunk contributions to geographic representations. Furthermore, we put forth an asynchronous update mechanism for the proposed task, aiming to guide the model to focus on specific chunks. Experiments on two Chinese benchmark datasets, show that the Geo-Encoder achieves significant improvements when compared to state-of-the-art baselines. Notably, it leads to a substantial improvement in the Hit@1 score of MGEO-BERT, increasing it by 6.22% from 62.76 to 68.98 on the GeoTES dataset.
M3 - Article in proceedings
SP - 1516
EP - 1530
BT - Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
PB - Association for Computational Linguistics (ACL)
Y2 - 17 March 2024 through 22 March 2024
ER -
ID: 385688033