NU_Hallucinators at NakbaArchiveClassifier Shared Task: A CLIP-Based Approach for Destruction Detection in Historical Image Archives

Salma Khaled Hegazy, Mohamed Ibrahim Ragab


Abstract
This paper presents a CLIP-based transfer learning approach for classifying historical archive images in the Nakba Image Classification Shared Task at the Nakba-NLP 2026 Workshop (LREC 2026). The task involves distinguishing images depicting destroyed or damaged infrastructure from those showing intact scenes using a dataset of 2,001 images collected from Instagram posts published by Palestinian content creators and journalists in Gaza between October 2023 and December 2025. Our method employs the CLIP ViT-B/32 visual encoder with selective fine-tuning of the final transformer block and a lightweight classification head. To address class imbalance, we apply focal loss along with standard data augmentation and threshold optimization. Experimental results show that the proposed model outperforms several CNN baselines and achieves an F1-score of 0.877 on the blind test set, securing 4th place in the shared task.
Anthology ID:
2026.nakbanlp-1.45
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:
289–293
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-nakbanlp-45
DOI:
10.63317/4c5nkazsww8o
Bibkey:
Cite (ACL):
Salma Khaled Hegazy and Mohamed Ibrahim Ragab. 2026. NU_Hallucinators at NakbaArchiveClassifier Shared Task: A CLIP-Based Approach for Destruction Detection in Historical Image Archives. In Proceedings of the 2nd International Workshop on Nakba Narratives as Language Resources @ LREC 2026, pages 289–293, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).
Cite (Informal):
NU_Hallucinators at NakbaArchiveClassifier Shared Task: A CLIP-Based Approach for Destruction Detection in Historical Image Archives (Hegazy & Ragab, NakbaNLP 2026)
Copy Citation: