Abstract: T-duality leads to a (continuous) O(d,d) symmetry of the tree-level closed string effective action reduced from D=10 to 10-d. However, this symmetry appears only after certain field redefinitions. I will argue that when the alpha'^3 correction is included these field redefinitions cannot be lifted to ten dimensions. This poses a problem for approaches like generalized geometry and double field theory, which assume an O(d,d) structure is present before reduction.
Link: Zoom
AnLy: https://sites.google.com/view/anlystringsandfields
