← All problems

Kuznetsov's categorical rationality conjecture for cubic fourfolds

16 years old · traced to

For a smooth cubic fourfold X⊂P5X\subset\mathbb P^5, there is a semiorthogonal decomposition

DbCoh⁡(X)=⟨Ku⁡(X),OX,OX(1),OX(2)⟩,D^b\operatorname{Coh}(X) =\langle\operatorname{Ku}(X),\mathcal O_X,\mathcal O_X(1),\mathcal O_X(2)\rangle,

where Ku⁡(X)\operatorname{Ku}(X) is a 22-Calabi-Yau category, often viewed as a noncommutative K3 surface.

Kuznetsov's conjecture. The fourfold XX is rational if and only if

Ku⁡(X)≃DbCoh⁡(S)\operatorname{Ku}(X)\simeq D^b\operatorname{Coh}(S)

for an ordinary, untwisted projective K3 surface SS.

There is a precise Hodge-theoretic characterization of when Ku⁡(X)\operatorname{Ku}(X) is equivalent to the derived category of a K3 surface, and recent work gives strong K3-type Hodge-theoretic necessary conditions for rationality. The full equivalence between rationality and geometricity of the Kuznetsov component remains open.

Progress summary

Refreshed
Open · progress reported

The conjecture remains open: established results characterize the geometric cases, but no reliable proof or counterexample to the rationality equivalence has appeared.

Kuznetsov formulated in 2008 the conjecture that a smooth cubic fourfold is rational exactly when its Kuznetsov component is equivalent to the derived category of an ordinary projective K3K3 surface.

Known results

  • Addington–Thomas (2012): geometricity of the Kuznetsov component is generically equivalent to Hassett’s Hodge-theoretic condition.
  • Bernardara–Macrì–Mehrotra–Stellari (2015): many generic components are equivalent to twisted K3K3 categories, not necessarily untwisted ones.
  • Bayer–Lahoz–Macrì–Nuer–Perry–Stellari (2024): geometricity is characterized by membership in an admissible Hassett divisor; rationality is known throughout C14\mathcal C_{14}, C26\mathcal C_{26}, and C38\mathcal C_{38}.

June 2026 Hodge-theoretic characterization

A 2026 result proves that having a Hodge-theoretically associated K3K3 surface is equivalent to AX≃Db(S)\mathcal A_X\simeq D^b(S), sharpening the categorical characterization without proving either rationality implication.

Community submission (unverified), October 8, 2026

A submitted argument claims that a manuscript produces irrational cubic fourfolds with geometric K3K3 categories in sufficiently large admissible Hassett divisors, which would refute the categorical-to-rational direction. No independent assessment corroborates the claim.

Current status (as of October 2026): The conjecture remains open; the Hodge-theoretic characterization and known rational families are established, while the alleged counterexample is unverified.

Sources

Solutions 1

Claimed by OpenAI.See full solutionHide full solution

The manuscript claims that a very general complex cubic fourfold in every sufficiently large admissible Hassett discriminant divisor is irrational while its Kuznetsov component is equivalent to the ordinary derived category of a projective K3 surface. The threshold is ineffective; this is a claimed counterexample to the categorical-to-rational direction of Kuznetsov’s conjecture.

Repository: https://github.com/openai/math

Manuscript: https://github.com/openai/math/blob/adc7f1241b42e322a6451854ab7e4b4c146bf78a/preprints/Irrational-cubic-fourfolds-with-geometric-K3-categories-September-24-2026/Irrational-cubic-fourfolds-with-geometric-K3-categories-September-24-2026.pdf

  • OpenAI-054-01-Irrational-cubic-fourfolds-with-geometric-K3-categories.pdf673,102 bytesOpen