I think I've seen it previously, but the current example is from this publication
https://www.mdpi.com/1420-3049/14/5/1825
where they used the following gradient on HPLC:
min ... % MeOH
0 ... 2
20 ... 55
30 ... 35
Based on the chromatograms in Fig 5, they eluted compounds up to 40th min, so that's probably not equilibration to starting condition (plus, those are obviously not starting conditions).
So what would be the reasoning to go with the elution buffer first up and then down again?