Generalized Fermat equation: a survey of solved cases (Q6962828)

In the paper the authors present a well-organized bibliographic survey of solved cases of the Diophantine equation \N\[\Nax^p+by^q=cz^r, \N\]\Nwhere \(a, b, c\in\mathbb{Z}\) and \(p, q, r \in\mathbb{N}_{\geq 2}\). Let us put \(\lambda(p,q,r)=1/p+1/q+1/r\). The triple \((p, q, r)\) is called the signature of the equation \(ax^p+by^q=cz^r\).\N\NThe authors present a classification by signature approach. In particular, if \(\lambda(p,q,r)>1\), then the corresponding curve (in an appropriate weighted projective space) is often parametrizable over \(\mathbb{Q}\). If \(\lambda(p,q,r)=1\) then we play with elliptic curves and have many various methods to find all primitive integer solutions. Finally, if \(\lambda(p,q,r)<1\) then the theorem of Darmon and Granville implies finiteness of primitive solutions. However, the proof of this theorem does not provide an effective method to determine the solutions.\N\NThe authors systematically collect what is known for many signatures and coefficient families. The summary tables are particularly useful: for instance, Table~1 compiles solved signatures in the Fermat-Catalan case \(x^p+y^q=z^r\) and highlights which ``smallest'' signatures remain open, helping readers avoid duplicating already-solved cases.\N\NThe way of presentation of the material is reference-driven rather than proof-driven and this is an intentional choice of the authors. Thus, this survey can be seen as a roadmap and entry point (especially for readers new to modular techniques), and as a practical checklist for researchers planning to attack new families. Therefore, the paper should serve as a convenient reference for the present state of the solvability of generalized Fermat equations.