The authors of this chapter present a two-fold approach for adapting content information delivered to a group of mobile users. This approach is based on a filtering process which considers both the user's current context and her/his preferences for this context. The authors propose an object-based context representation, which takes into account the user's physical and collaborative contexts, including elements related to collaboration tasks and group work in which the user is involved. They define the notion of preference for an individual or a group of people that develops a collaborative task and give a typology of preferences before proposing a formalism to represent them. This representation is exploited by a context matching algorithm in order to select only user preferences which can be applied according to the context of use. This chapter also presents the framework PUMAS which adopts a Multi-Agent System approach to support our propositions.