Revalidation Beats Stateful Routing for Scientific Surrogates Under Distribution Shift
Quick summary
arXiv:2609.29715v1 Announce Type: cross Abstract: Surrogate models are often chosen during development and then left in place as new measurements arrive. That practice becomes risky when noise, input support, or physical parameters change. We asked whether such changes call for a stateful adaptive controller, or whether it is enough to validate the candidate models again on each new batch. To study this question, we built RegimeShift-Surrogates, a reproducible streaming benchmark spanning eight analytic and dynamical tasks, four stationary or shifting regimes, ten held-out seeds, and eight cla
Key takeaways
- arXiv:2609.29715v1 Announce Type: cross Abstract: Surrogate models are often chosen during development and then left in place as new measurements arrive.
- That practice becomes risky when noise, input support, or physical parameters change.
- We asked whether such changes call for a stateful adaptive controller, or whether it is enough to validate the candidate models again on each new batch.
Why it matters
“Revalidation Beats Stateful Routing for Scientific Surrogates Under Distribution Shift” highlights the need for repeatable measurement rather than a single impressive demonstration. Independent validation across datasets and clearly stated limitations determine whether a result can guide product decisions.
Member comments