New arXiv paper proposes a distributionally robust framework for survival analysis that jointly handles latent subpopulation shift and outlier contamination via outer minimization and inner maximization. Relevant for AI…

#SurvivalAnalysis #RobustML #DistributionAI
https://arxiv.org/abs/2610.02868

Time series forecasting benchmarks often miss real-world failure modes by relying on Gaussian noise or simple perturbations instead of structured, scenario-grounded stress tests. The rise of TSF…

#TimeSeriesForecasting #Benchmarking #AIInfrastructure #RobustML
https://arxiv.org/abs/2610.02608

If you are attending @bmvcconf.bsky.social, come and listen about open-set segmentation in medical imaging from the great Luisa Vargas.

When & Where? Nov 25th, 16h15 GMT - Main Hall

In the meantime, check the paper (bmva-archive.org.uk/bmvc/2025/as...) and the code (github.com/robustml-eur...).

Researchers propose an iterative training algorithm for Neural ODEs using gradient descent in a Banach subspace, showing reduced performance loss under simulated disturbances. Read more: https://getnews.me/training-method-boosts-neural-odes-resistance-to-disturbances/ #neuralode #robustml

#RobustLLMs is part of the #ELLIS RobustML Programme and takes place at Keble College, Oxford. It is hosted by the Department of Statistics at Oxford and co-organized by the ELISE network.

More information about the workshop can be found here: sites.google.com/view/robustm...

Joint work with @blackhc.bsky.social @thoefler.bsky.social @arkrause.bsky.social @nmervegurel.bsky.social
accepted at #AISTATS2025 !

Code: github.com/RobustML-Lab...
Paper: arxiv.org/abs/2410.13609