Minimizing risk through minimizing model-data interaction: a protocol for relying on proxy tasks when designing child sexual abuse imagery detection models (2025)
- Authors:
- Autor USP: RIBEIRO, LEO SAMPAIO FERRAZ - ICMC
- Unidade: ICMC
- DOI: 10.1145/3715275.3732102
- Subjects: RECONHECIMENTO DE IMAGEM; APRENDIZADO COMPUTACIONAL; ABUSO SEXUAL; CRIANÇAS
- Keywords: Child Sexual Abuse Recognition; Few-shot Learning; Scene Classification
- Agências de fomento:
- Language: Inglês
- Objetivos de Desenvolvimento Sustentável (ODS):
16. Paz, justiça e instituições eficazes
- Imprenta:
- Source:
- Título: Proceedings
- Conference titles: ACM Conference on Fairness, Accountability, and Transparency - FAccT
- Este periódico é de acesso aberto
- Este artigo NÃO é de acesso aberto
-
ABNT
COELHO, Thamiris Florindo et al. Minimizing risk through minimizing model-data interaction: a protocol for relying on proxy tasks when designing child sexual abuse imagery detection models. 2025, Anais.. New York: ACM, 2025. Disponível em: https://doi.org/10.1145/3715275.3732102. Acesso em: 20 jan. 2026. -
APA
Coelho, T. F., Ribeiro, L. S. F., Macedo, J., Santos, J. A. dos, & Avila, S. (2025). Minimizing risk through minimizing model-data interaction: a protocol for relying on proxy tasks when designing child sexual abuse imagery detection models. In Proceedings. New York: ACM. doi:10.1145/3715275.3732102 -
NLM
Coelho TF, Ribeiro LSF, Macedo J, Santos JA dos, Avila S. Minimizing risk through minimizing model-data interaction: a protocol for relying on proxy tasks when designing child sexual abuse imagery detection models [Internet]. Proceedings. 2025 ;[citado 2026 jan. 20 ] Available from: https://doi.org/10.1145/3715275.3732102 -
Vancouver
Coelho TF, Ribeiro LSF, Macedo J, Santos JA dos, Avila S. Minimizing risk through minimizing model-data interaction: a protocol for relying on proxy tasks when designing child sexual abuse imagery detection models [Internet]. Proceedings. 2025 ;[citado 2026 jan. 20 ] Available from: https://doi.org/10.1145/3715275.3732102 - Cross Domain Visual Search with Feature Learning using Multi-stream Transformer-based Architectures
- Neglected risks: the disturbing reality of children's images in datasets and the urgent call for accountability
- Scene Designer: a unified model for scene search and synthesis from sketch
- Scene designer: compositional sketch-based image retrieval with contrastive learning and an auxiliary synthesis task
- Sketchformer: transformer-based representation for sketched structure
- Proxy tasks ensemble for explainable inference in sensitive data
- Training deep networks from zero to hero: avoiding pitfalls and going beyond
Informações sobre o DOI: 10.1145/3715275.3732102 (Fonte: oaDOI API)
How to cite
A citação é gerada automaticamente e pode não estar totalmente de acordo com as normas
