On the finiteness of \(\text{ Ш}\) for motives associated to modular forms (Q1356000)

Let \(f\) be a new form of even weight \(2r\) for \(\Gamma_0(N)\), and let \({\mathcal M}_f\) be the \(r\)-th Tate twist of the motive associated to \(f\). Let \(K\) be a quadratic imaginary field of discriminant \(D\) in which all primes dividing \(N\) split, and let \(p\) be a prime not dividing \(2N\). Let \(T_p({\mathcal M}_f)\) be the \(p\)-adic realization of \({\mathcal M}_f\) and let \(P(1)\) be the image in \(H^1(K,T_p({\mathcal M}_f))\) of the Heegner cycle associated with \(K\) under the \(p\)-adic Abel-Jacobi map. The main result of the paper is the following theorem, generalizing a result of Nekovář: There is a finite set of primes \(\Psi(f)\), depending only on \(f\), such that for a prime \(p\not\in\Psi(f)\) the following holds: With \(K\) as above, if \(P(1)\) is not torsion, then \(p^{2{\mathcal I}_p}\text{X}_p({\mathcal M}_f/K)=0\), where \({\mathcal I}_p\) is the smallest non-negative integer such that the reduction of \(P(1)\) to \(H^1(K,T_p({\mathcal M}_f)/p^{{\mathcal I}_p+1})\) is not \(0\). In particular, if \({\mathcal I}_p=0\), then \(\text{X}_p({\mathcal M}_f/K)=0\). Let \(f\) have the \(q\)-expansion \(f=\sum a_nq^n\) and let \(E={\mathbb Q}(a_n)\) be the field generated over \({\mathbb Q}\) by the \(a_n\). For a prime \(p\), let \(\mathfrak p\) be a prime in \(E\) over \(p\), and let \(\rho_{f,p}\) be the \(p\)-adic representation associated with \(f\). The \(\mathfrak p\) component of \(\rho_{f,p}\) is a representation of \(\text{ Gal}(\overline{\mathbb Q}/{\mathbb Q})\) on a 2-dimensional \(E_{\mathfrak p}\)-vector space \(V_{\rho_f,\mathfrak p}\). Let \(S=\{g\in\text{ GL}_2({\mathbb Z}_p)|\det g\in(({\mathbb Z}_p^{\times})^{2r-1})\}\). The set \(\Psi(f)\) is now the set containing the primes dividing \(2N\), primes that ramify in \(E\) and primes where the image of \(\text{ Gal}(\overline{\mathbb Q}/{\mathbb Q})\) in \(\text{ Aut}(V_{\rho_f,\mathfrak p})\) does not contain \(S\). The proof of the main result uses a variant of the Kolyvagin argument as in [\textit{B. Gross}, in: \(L\)-functions and arithmetic, Lond. Math. Soc. Lect. Note Ser. 153, 235-256 (1991; Zbl 0743.14021)].