Developing self-supervised learning (SSL) models that can learn universal and transferable representations of H&E gigapixel whole-slide images (WSIs) is becoming increasingly valuable in computational pathology. These models hold the potential to advance critical tasks such as few-shot classification, slide retrieval, and patient stratification. Existing approaches for slide representation learning extend the principles of SSL from small images (e.g., 224 x 224 patches) to entire slides, usually by aligning two different augmentations (or views) of the slide. Yet the resulting representation remains constrained by the limited clinical and biological diversity of the views. Instead, we postulate that slides stained with multiple markers, such as immunohistochemistry, can be used as different views to form a rich task-agnostic training signal. To this end, we introduce Madeleine, a multimodal pretraining strategy for slide representation learning. Madeleine is trained with a dual global-local cross-stain alignment objective on large cohorts of breast cancer samples (N=4,211 WSIs across five stains) and kidney transplant samples (N=12,070 WSIs across four stains). We demonstrate the quality of slide representations learned by Madeleine on various downstream evaluations, ranging from morphological and molecular classification to prognostic prediction, comprising 21 tasks using 7,299 WSIs from multiple medical centers. Code is available at https://github.com/mahmoodlab/MADELEINE.

Data scarcity has been a long standing issue in the field of open-domain social dialogue. To quench this thirst, we present SODA: the first publicly available, million-scale high-quality social dialogue dataset. By contextualizing social commonsense knowledge from a knowledge graph, we are able to distill an exceptionally broad spectrum of social interactions from a large language model. Human evaluation shows that conversations in SODA are more consistent, specific, and (surprisingly) natural than those in prior human-authored datasets. Using SODA, we train COSMO: a generalizable conversation model that is significantly more natural and consistent on unseen datasets than best-performing conversation models (e.g., GODEL, BlenderBot-1, Koala, Vicuna). Experiments reveal COSMO is sometimes even preferred to the original human-written gold responses. Additionally, our results shed light on the distinction between knowledge-enriched conversations and natural social chitchats. We plan to make our data, model, and code public.

THE TVET EXPERT OF THE WEEK, p18 Madeleine Anne Decker is information and knowledge management specialist for the Canadian Vocational Association/Association canadienne de la formation professionnelle. The Canadian Vocational Association (CVA) was created in 1960 to promote and foster education and training which leads to occupational competence. The CVA is also a world leader in DACUM training, development and research, and Canada's premier voice and expert in competency-based learning and training. Madeleine joined the CVA team in 2011 to increase the visibility by publishing a monthly newsletter and creating two TVET online databases. She added to these tasks the role of community manager by feeding the Association's Twitter and LinkedIn accounts.

To us as a neuroscientist and biomechanist ( Lena), and a rehabilitation scientist and dancer ( Madeleine), understanding the complexities of motor skill in a ballet move, or the physical language of coordination in partner dance, is an inspiring and daunting challenge. Adapted tango rehabilitation class improves gait and balance in people with Parkinson's disease. Lucas McKay, an assistant professor in Biomedical Engineering at Emory specializing in mechanisms of balance impairment in Parkinson's disease, showed that participants improved muscle activity for balance after adapted tango. That is, as they practiced their tango dancing skills, they developed motor modules that also helped them walk and balance in everyday situations.