| Paper: Gradient Flow Sampler-based Distributionally Robust Optimization | |
| OpenReview: QRtzkKrbJi | |
| ArXiv: 2510.25956 | |
| Literal registry source: arXiv v1 | |
| v1 PDF SHA-256: 796d3cb25e2a9062ab4e81201daa7837c4bf615399063b0ed439c9e17e6db8d6 | |
| v1 source SHA-256: 35a471bd60c11517db90b8e337064019c5a07faa621718aeca9800d425604488 | |
| current PDF SHA-256: 88720453447d8affcefd3e63097c52f512d5b275766ba3e73dff9af2ca1c0f8f | |
| current source SHA-256: cc00b5195a2592b24f9b876bf9fd27ada78fa9a46c468e81221983943cb726e5 | |
| V1 exactly matches the registered O(epsilon_opt^-2) outer-loop and O-tilde(epsilon_opt^-4) WGF total-complexity claims. The current revision changes these to epsilon_opt^-4 and epsilon_opt^-6 and is a disclosed drift control only. | |