The field of bioacoustics is moving towards the development of more advanced and generalizable models, with a focus on improving zero-shot generalization and uncertainty calibration. Researchers are exploring new methods for passive acoustic monitoring, including the use of denoising techniques to improve the accuracy of reef health assessments. The importance of evaluating and improving uncertainty calibration in bioacoustic classifiers is also being highlighted, with studies demonstrating the effectiveness of simple post hoc calibration methods. Furthermore, there is a growing interest in understanding the development of visual expertize in humans, with the introduction of large-scale knowledge tracing benchmarks for fine-grained bird species recognition. Noteworthy papers include: Model Merging Improves Zero-Shot Generalization in Bioacoustic Foundation Models, which demonstrates a significant improvement in zero-shot generalization using a simple model merging strategy. Passive Acoustic Monitoring of Noisy Coral Reefs, which shows that passive acoustics can be used to monitor reef health, provided the data is effectively denoised and interpreted. Uncertainty Calibration of Multi-Label Bird Sound Classifiers, which highlights the importance of evaluating and improving uncertainty calibration in bioacoustic classifiers.