Fiona Margaret Callahan

STEM-LM uses masked language modeling to predict multi-species distributions by treating species presence as tokens in a spatio-temporal 'sentence', achieving high discrimination on ecological datasets.

This study shows that co-occurrence networks can retain key features of species interaction networks, suggesting they can be used to test hypotheses about ecological community dynamics.

The paper evaluates five methods for detecting ecological interactions from sedimentary ancient DNA data and finds high false discovery rates and poor detection performance under realistic conditions.