Canonical vertex formalism in DT theory of toric Calabi-Yau \(4\)-folds (Q2078518)

Donaldson-Thomas invariants are virtual counts of stable sheaves on Calabi-Yau 3-folds [\textit{R. P. Thomas}, J. Differ. Geom. 54, No. 2, 367--438 (2000; Zbl 1034.14015)]. It has been recently generalized to a theory on Calabi-Yau 4-folds [\textit{Y. Cao} and \textit{N. C. Leung}, ``Donaldson-Thomas theory for Calabi-Yau 4-folds'', Preprint, \url{arXiv:1407.7659}; \textit{D. Borisov} and \textit{D. Joyce}, Geom. Topol. 21, No. 6, 3231--3311 (2017; Zbl 1390.14008); \textit{J. Oh} and \textit{R. P. Thomas}, ``Counting sheaves on Calabi-Yau 4-folds. I'', Preprint, \url{arXiv:2009.05542}]. One key input of the theory is a choice of orientation whose existence is proved by [\textit{Y. Cao} et al., Adv. Math. 368, Article ID 107134, 59 p. (2020; Zbl 1443.14056)] for projective Calabi-Yau 4-folds. Using the same method, the orientability result is extended to quasi-projective Calabi-Yau 4-folds by [\textit{A. Bojko}, Adv. Math. 388, Article ID 107859, 59 p. (2021; Zbl 1478.14082)]. But unfortunately, it is quite difficult to give a canonical choice of orientations, although orientations constructed in [loc. cit.] are given by the pullback of orientations from connected gauge theoretic moduli spaces whose choices become ``two''. DT theory on Calabi-Yau 4-folds has deep connections to many other subjects: Gromov-Witten theory, combinatorics of solid partitions and SUSY gauge theory, see e.g. [\textit{Y. Cao} et al., Adv. Math. 338, 41--92 (2018; Zbl 1408.14177); \textit{Y. Cao} et al., J. Eur. Math. Soc. (JEMS) 24, No. 2, 527--581 (2022; Zbl 1491.14077); \textit{Y. Cao} and \textit{M. Kool}, Adv. Math. 338, 601--648 (2018; Zbl 1420.14126); \textit{N. Nekrasov} and \textit{N. Piazzalunga}, Commun. Math. Phys. 372, No. 2, 573--597 (2019; Zbl 07136912)]. In [loc. cit.], the choice of orientation is based on case by case study and it seems difficult to link those orientations with orientations coming from gauge theoretic moduli spaces. One important case which has been extensively studied is when the Calabi-Yau is toric and one can then study invariants at the level of fixed points (by vertex formalism). The advantage is that everything becomes computable but the tricky point is one has to decide a choice of sign at each torus fixed point. In this paper under review, the author gives a conjectural choice of signs in above story, which generalizes the sign rule of Nekrasov-Piazzalunga. He also showed the canonicity of the signs and compatibility with signs rules in computations in those papers mentioned above. The paper is an important reference for anyone who wants to do computations in DT4 theory using vertex formalism.