An ultimate proof of Hoffmann-Totaro's conjecture (Q2098202)

Given an anisotropic quadratic form \(\varphi\) over a field \(F\), the restriction \(\varphi_K\) is isotropic where \(K\) is the function field of its underlying quadric. The maximal dimension of a totally isotropic subform, denoted by \(i_1(\varphi)\), was conjectured by Hoffmann (originally in characteristic 2) and Totaro (arbitrary characteristic, see [\textit{B. Totaro}, J. Algebr. Geom. 17, No. 3, 577--597 (2008; Zbl 1144.11031), p. 596]) to be at most \(2^m\) where \(m\) is the the maximal integer for which \(2^m | \dim \varphi-i_1(\varphi)\). The conjecture implies, in particular, that \(i_1(\varphi)\leq \frac{1}{2} \dim \varphi\), something that is well known in the nonsingular case (where \(i_1(\varphi)\) coincides with the first Witt index), and the bound is sharp when \(\varphi\) is a Pfister form. The conjecture was proven by the author of this current paper to hold true when \(\operatorname{char}(F)\neq 2\) in [\textit{N. A. Karpenko}, Invent. Math. 153, No. 2, 455--462 (2003; Zbl 1032.11016)], using Steenrod operations on modulo 2 Chow groups. In the works of Primozic, Haution and Scully, much of the machinery was adapted to the characteristic 2 case, and the conjecture remained open only in the case of singular, but not totally singular, quadratic forms over fields of characteristic 2, a case which the current paper resolves in the positive (as expected).