KvochurHegel at NakbaArchiveClassifier Shared Task: Nakba Image Classification via ConvNeXt-V2 and Label Smoothing

Minh-Hoang Le


Abstract
This paper presents the KvochurHegel team’s submission to the Nakba Image Classification shared task at the Nakba-NLP 2026 Workshop. The task requires the binary classification of social media images into destruction and not_destruction categories. Given a limited and imbalanced training set of 1,400 images, we utilized a ConvNeXt-V2 Nano backbone combined with extensive data augmentation and label smoothing, prioritizing standard regularization over task-specific architectural modifications. For inference, we applied a 6-view Test-Time Augmentation (TTA) strategy using a hard-voting mechanism. The baseline system achieved a Macro F1-score of 0.8593 and an Accuracy of 0.8706 on the official private test set, ranking 6th out of 16 participating teams.
Anthology ID:
2026.nakbanlp-1.14
Volume:
Proceedings of the 2nd International Workshop on Nakba Narratives as Language Resources @ LREC 2026
Month:
May
Year:
2026
Address:
Palma, Mallorca (Spain)
Editors:
Mustafa Jarrar, Mo El-Haj, Amal Haddad, Serin Atiani, Shadi Abudalfa, Terry Regier, Paul Rayson, Khalil Sima’an, Camille Mansour
Venues:
NakbaNLP | WS
SIG:
Publisher:
ELRA Language Resources Association (ELRA)
Note:
Pages:
118–120
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-nakbanlp-14
DOI:
10.63317/4uecz9j2m37s
Bibkey:
Cite (ACL):
Minh-Hoang Le. 2026. KvochurHegel at NakbaArchiveClassifier Shared Task: Nakba Image Classification via ConvNeXt-V2 and Label Smoothing. In Proceedings of the 2nd International Workshop on Nakba Narratives as Language Resources @ LREC 2026, pages 118–120, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).
Cite (Informal):
KvochurHegel at NakbaArchiveClassifier Shared Task: Nakba Image Classification via ConvNeXt-V2 and Label Smoothing (Le, NakbaNLP 2026)
Copy Citation: