VibeMathedMath problems solved with AI

Darboux injections from closed manifolds: Banakh–Banakh Problems 1.7 and 1.8

Banakh and Banakh (2020) proved that connectedness-preserving (Darboux) injections are continuous in several compact settings — into 1-manifolds from compact sources, from closed surfaces into surfaces, and from closed 3-manifolds with finite H1H_1 into 3-manifolds — and asked whether every Darboux bijection of S4\mathbb S^4 (Problem 1.7) and of T3\mathbb T^3 (Problem 1.8) is a homeomorphism. Answer: yes. For every n2n\ge2, every Darboux injection from a connected closed nn-manifold into an nn-manifold is a homeomorphism onto a connected component of the target; no homology hypothesis and no surjectivity are needed

Result
Proved(see note)
Status
Candidate (review pending)
AI contribution
AI-discovered
Method
Argument
Field
Continuum theory / Darboux maps
Posed by
Iryna Banakh and Taras Banakh, Topology and its Applications 275 (2020), Problems 1.7–1.8
Year posed
2020
Years open
6y
Solved
2026-06
Model
GPT-5.5 Pro
Vendor
OpenAI
Collaborators
Peter L.
Verification
Unreviewed
Publication
Preprint
Significance
15 / 100
Disclosed cost
Wikipedia
No dedicated article

What was actually shown

Both problems are answered affirmatively, in every dimension at once: every connectedness-preserving injection from a connected closed nn-manifold into an nn-manifold is a homeomorphism onto a component, so in particular every Darboux self-bijection of Sn\mathbb S^n and of every closed manifold is a homeomorphism. This removes the finite-H1H_1 hypothesis of the 2020 theorem for 3-manifolds and extends it above dimension 3. Compactness of the source is essential: the companion preprint (Zenodo 10.5281/zenodo.22346412) shows the corresponding statement fails for Rn\mathbb R^n, n2n\ge2. The note does not address noncompact sources, manifolds with boundary, or targets of different dimension.

What the AI did

The note's acknowledgment: it was produced with substantial assistance from large language models, principally GPT-5.5 Pro, in a research program directed by the author, who selected, checked and assembled the arguments. The proof uses Banakh–Banakh's framework of nn-varieties and componnectedness, Alexander–Lefschetz duality with F2\mathbb F_2 coefficients, and an induction on minimal carriers of nonzero Čech cohomology classes; a separate proof that metrizable nn-manifolds are nn-varieties, and a one-dimensional base case, are supplied. The same theorem was later re-derived by the same method, independently and without access to the note, by GPT-6 (Codex) in a subsequent phase of the program; that re-derivation is in the program's records.

Verification

Unreviewed. The 17-page note (Zenodo 10.5281/zenodo.22347647, dated June 2026, posted 5 September) was read here in full; the theorem, the method (Alexander–Lefschetz duality with F2 coefficients, induction on minimal carriers of Čech cohomology classes, in Banakh–Banakh's framework of n-varieties) and the disclosure match the submission, and Problems 1.7 and 1.8 were confirmed verbatim in arXiv 1809.00401. Nobody outside the author's program has read the argument; the re-derivation by a second model inside the same program is internal corroboration. Candidate as submitted.

Sources

Submitted by WittyHeron892 on

Changelog2 changes

Discussion