Twente-BMS-NLP at PerspectiveArg 2024: Combining Bi-Encoder and Cross-Encoder for Argument Retrieval

Leixin Zhang, Daniel Braun


Abstract
The paper describes our system for the Perspective Argument Retrieval Shared Task. The shared task consists of three scenarios in which relevant political arguments have to be retrieved based on queries (Scenario 1). In Scenario 2 explicit socio-cultural properties are provided and in Scenario 3 implicit socio-cultural properties within the arguments have to be used. We combined a Bi-Encoder and a Cross-Encoder to retrieve relevant arguments for each query. For the third scenario, we extracted linguistic features to predict socio-demographic labels as a separate task. However, the socio-demographic match task proved challenging due to the constraints of argument lengths and genres. The described system won both tracks of the shared task.
Anthology ID:
2024.argmining-1.17
Volume:
Proceedings of the 11th Workshop on Argument Mining (ArgMining 2024)
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Yamen Ajjour, Roy Bar-Haim, Roxanne El Baff, Zhexiong Liu, Gabriella Skitalinskaya
Venue:
ArgMining
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
164–168
Language:
URL:
https://aclanthology.org/2024.argmining-1.17
DOI:
10.18653/v1/2024.argmining-1.17
Bibkey:
Cite (ACL):
Leixin Zhang and Daniel Braun. 2024. Twente-BMS-NLP at PerspectiveArg 2024: Combining Bi-Encoder and Cross-Encoder for Argument Retrieval. In Proceedings of the 11th Workshop on Argument Mining (ArgMining 2024), pages 164–168, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Twente-BMS-NLP at PerspectiveArg 2024: Combining Bi-Encoder and Cross-Encoder for Argument Retrieval (Zhang & Braun, ArgMining 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.argmining-1.17.pdf