English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Foundation models are a current focus of attention in both industry andacademia. While they have shown their capabilities in a variety of tasks,in-depth research is required to determine their robustness to distributionshift when used as a basis for supervised machine learning. This is especiallyimportant in the context of clinical data, with particular limitations relatedto data accessibility, lack of pretraining materials, and limited availabilityof high-quality annotations. In this work, we examine the stability of modelsbased on representations from foundation models under distribution shift. Wefocus on confounding by provenance, a form of distribution shift that emergesin the context of multi-institutional datasets when there are differences insource-specific language use and class distributions. Using a sampling strategythat synthetically induces varying degrees of distribution shift, we evaluatethe extent to which representations from foundation models result inpredictions that are inherently robust to confounding by provenance.Additionally, we examine the effectiveness of a straightforward confoundingadjustment method inspired by Pearl's conception of backdoor adjustment.Results indicate that while foundation models do show some out-of-the-boxrobustness to confounding-by-provenance related distribution shifts, this canbe considerably improved through adjustment. These findings suggest a need fordeliberate adjustment of predictive models using representations fromfoundation models in the context of source-specific distributional differences.

Enhancing Robustness of Foundation Model Representations under Provenance-related Distribution Shifts