Abstract: Entity resolution (ER), a longstanding problem of data cleaning and integration, aims at identifying different data records that represent the same real-world entity. Existing approaches treat ER focus only on finding perfectly matched records and separating the corresponding from non-corresponding ones. However, in real-world scenarios, where ER is part of a more general data project, downstream applications may not only require resolution of records that refer to the same entity but may also seek to match records that share different levels of commonality, relating, for example, to various granularity levels of the resolution. In what follows, we introduce the problem of multiple intents entity resolution (MIER), an extension to the universal (single intent) ER task. As a solution, we propose FlexER, utilizing contemporary solutions to universal ER tasks to solve multiple intents entity resolution. FlexER addresses the problem as multi-label classification and combines intent-based representations of record pairs using a graph convolutional network (GCN) to improve the outcome to multiple resolution problems. A large-scale empirical evaluation introduces a new benchmark and, using also three well-known benchmarks, shows that FlexER effectively solves the MIER problem and outperforms the state-of-the-art for a universal ER.