Rivière’s regularity question for critical $n$-Laplace systems with antisymmetric potentials
resolvedconfidence 70%
VibeMathed reports this item as resolved. VibeMath preserves that report as a source assertion and has not independently authored a plain-language mathematical explanation.
Precise statement
Let $n>2$. We construct a map $U\in W^{1,n}(B^n,\mathbb{R}^{n+2})$ that is discontinuous at the origin and smooth on the punctured ball $B^n \setminus \{0\}$, together with an antisymmetric potential $\Omega\in L^n(B^n,so(n+2)\otimes\mathbb{R}^n)$ such that $-\mathrm{Div}(|\nabla U|^{n-2}\nabla U)=\Omega\cdot |\nabla U|^{n-2}\nabla U$ in $D'(B^n)$. This gives a negative answer to a regularity question posed by Rivière.
Our potential admits the Lorentz-space regularity $\Omega \in \bigcap_{q>2}L^{(n,q)} \setminus L^{(n,2)}$. In addition for given $1<p<\infty$ we can enforce $\nabla U \in L^{(n,p)}$ but $\nabla U \notin L^{(n,1)}$. The construction does not give a counterexample to regularity for weakly $n$-harmonic maps or for higher-dimensional $H$-systems.
The example was generated by ChatGPT 5.6 Sol on August 5, 2026. The work itself was written by the author and thoroughly reviewed to ensure its correctness.
The source statement is reproduced for indexing with attribution. Mathematical correctness requires domain-expert or mechanical review. VibeMath has not independently audited statement
fidelity, correctness, priority, or novelty.
What AI did
ChatGPT 5.6 Sol
From the paper's own "AI Usage" section, which also appears in condensed form on the title page: "The example was provided by ChatGPT 5.6 Sol while the author was exploring possible counterexamples to the regularity problem of weakly $n$-harmonic maps on August 5, 2026. The author identified it as a solution to the more general problem with the antisymmetric potential as in (1.1). Furthermore, the author simplified the paramters and notations for better readability. The proof has been checked by the author and is correct. The work was written by the author, however code snippets may occasionally come from LLMs including ChatGPT 5.6 Sol or Gemini 3.6 Thinking." For a disproof the counterexample is the entire result, and the model produced it; the author's contributions as he describes them are recognising what it settled, simplifying the parameters, checking the proof and writing the paper. Hence AI-discovered.
An arXiv preprint (v1, 25 August 2026, math.AP), unrefereed and with no independent endorsement, and no mathematics was checked here - there is no formalization and no computational certificate. The author states he has checked the proof himself, which is his own assurance rather than an independent one. Verified here on 26 August 2026: the paper exists at arXiv:2608.24393 with the title, sole author and construction this entry describes; its AI-usage section carries the disclosure quoted above word for word; and the question it answers is real and traceable - the abstract cites Rivière's 2011 chapter "The role of integrability by compensation in conformal geometric analysis" (Séminaires et Congrès 22) at Eq. (3.23), reformulated as open Problem 2.5 in Schikorra and Strzelecki's 2017 EMS survey on H-systems in higher dimensions.
For every $n>2$, the paper constructs a bounded map $U\in W^{1,n}(B^n,\mathbb{R}^{n+2})$, smooth on $B^n\setminus\{0\}$ but discontinuous at the origin, together with an antisymmetric potential
$$
\Omega\in L^n(B^n,so(n+2)\otimes\mathbb{R}^n)
$$
such that
$$
-\mathrm{Div}\bigl(|\nabla U|^{n-2}\nabla U\bigr)
=
\Omega\cdot|\nabla U|^{n-2}\nabla U
\qquad\text{in }D'(B^n).
$$
Moreover, the potential satisfies the sharper Lorentz-space regularity
$$
\Omega\in\bigcap_{q>2}L^{(n,q)}\setminus L^{(n,2)},
$$
and, for any prescribed $1<p<\infty$, the construction can be arranged so that
$$
\nabla U\in L^{(n,p)}
\qquad\text{but}\qquad
\nabla U\notin L^{(n,1)}.
$$
This gives a negative answer to Rivière’s general regularity question for critical $n$-Laplace systems with antisymmetric $L^n$ potentials: antisymmetry and critical $L^n$ control alone do not imply continuity.
Known method families
construction (source-reported)
Source-reported tools: construction.
Independent: unknown · difference confidence: 0
What remains uncertain
VibeMath has not independently audited the mathematical statement, proof, or novelty claim.
VibeMath has not independently verified the mathematical claim.
AI-attempt independence and training-data exposure are unknown unless explicitly documented.
VibeMath has not independently audited the mathematical statement, proof, or novelty claim.