André C. Ferreira

This paper introduces an LSTM-based framework for automated nest behaviour recognition in wild birds, outperforming human annotators and YOLO in accuracy and speed.

The study investigates sentinel behavior in sociable weavers using an artificial setup and finds no evidence of this behavior occurring.

This study examines infanticide in sociable weavers and finds no direct reproductive benefits, suggesting indirect fitness benefits may drive the behavior.

The paper examines sex-specific breeding and helping strategies in sociable weavers, finding males help more frequently and for longer periods than females, with notable role-switching behavior.

Across eleven years and more than 1,800 sociable weavers, black bib size grew with age but did not predict dominance; age predicted dominance in males only.

This paper argues for moving beyond performance metrics in ML classification for animal studies and proposes simulations and biological validations to better assess model utility despite noisy data.

The paper experimentally shows that increased foraging encounters with non-kin promote social associations in a kin-based bird society.

This paper uses deep learning to detect sexual dimorphism in sociable weavers, finding higher accuracy than humans and identifying beak regions as key for sex prediction.

The paper investigates whether social associations between breeders and helpers in sociable weavers are linked to the help received, finding stronger foraging bonds and positive correlations between social affiliations and help provided.