← All problems

Maximization of the Yamabe invariant by quotient-round lens-space metrics

For every odd integer n≥5n\ge 5 and every positive integer mm, let Lmn=Sn/CmL_m^n=S^n/C_m be the standard lens space and let [grd][g_{\rm rd}] be its quotient-round conformal class. The conjecture asserts that σ(Lmn)=Y(Lmn,[grd])=m−2/nY(Sn)\sigma(L_m^n)=Y\bigl(L_m^n,[g_{\rm rd}]\bigr)=m^{-2/n}Y(S^n); equivalently, the quotient-round conformal class maximizes the smooth Yamabe invariant among all conformal classes on LmnL_m^n.

References

Source: arXiv

Progress summary

Refreshed
Open · solution reported

A preprint claims a counterexample in odd dimensions five and higher, so quotient-round metrics do not always maximize the smooth Yamabe invariant.

The problem asks whether quotient-round metrics maximize the smooth Yamabe invariant for lens spaces in odd dimensions. No proposer or original date is identified in the retrieved sources.

October 13, 2026 counterexample

Jakob Loren's preprint claims that for L3qn=Sn/C3qL_{3q}^n=S^n/C_{3q}, with odd n≥5n\ge 5 and prime q>3q>3, sufficiently large qq satisfy

σ(L3qn)>(3q)−2/nY(Sn)=Y(L3qn,[grd]).\sigma(L_{3q}^n)>(3q)^{-2/n}Y(S^n)=Y\bigl(L_{3q}^n,[g_{\mathrm{rd}}]\bigr).

Thus the quotient-round conformal class does not maximize the invariant for this infinite family. It also claims the lower bound σ(L3qn)≥min⁡{q−2/nY(Sn),Λn}\sigma(L_{3q}^n)\ge\min\{q^{-2/n}Y(S^n),\Lambda_n\}, derived using spin bordism and surgery. The claim is unverified.

Current status (as of October 2026): A preprint claims the universal maximization expectation is false for an infinite family with odd n≥5n\ge 5 and sufficiently large prime qq, while the assertion for individual lens spaces outside that family remains open.

Sources

Solutions 0

No solutions have been posted yet.