Modern models of relation extraction for tasks like ACE are based on supervised learning of relations from small hand-labeled corpora. We investigate an alternative paradigm that does not require labeled corpora, avoiding the domain dependence of ACEstyle algorithms, and allowing the use of corpora of any size. Our experiments use Freebase, a large semantic database of several thousand relations, to provide distant supervision. For each pair of entities that appears in some Freebase relation, we find all sentences containing those entities in a large unlabeled corpus and extract textual features to train a relation classifier. Our algorithm combines the advantages of supervised IE (combining 400,000 noisy pattern features in a probabilistic classifier) and unsupervised IE (extracting large numbers of relations from large corpora of any domain). Our model is able to extract 10,000 instances of 102 relations at a precision of 67.6%. We also analyze feature performance, showing that syntactic parse features are particularly helpful for relations that are ambiguous or lexically distant in their expression.

We describe an access control model that has been implemented in the web content management framework "Deme" (which rhymes with "team"). Access control in Deme is an example of what we call "bivalent relation object access control" (BROAC). This model builds on recent work by Giunchiglia et al. on relation-based access control (RelBAC), as well as other work on relational, flexible, fine-grained, and XML access control models. We describe Deme's architecture and review access control models, motivating our approach. BROAC allows for both positive and negative permissions, which may conflict with each other. We argue for the usefulness of defining access control rules as objects in the target database, and for the necessity of resolving permission conflicts in a social Web/collaboration architecture. After describing how Deme access control works, including the precedence relations between different permission types in Deme, we provide several examples of realistic scenarios in which permission conflicts arise, and show how Deme resolves them. Initial performance tests indicate that permission checking scales linearly in time on a practical Deme website.

Abstract-This paper addresses the problem of coordinating great numbers of vehicles in large geographical areas under network connective constraints. We leverage previous work on hierarchical potential fields to create advanced skills in multirobot systems. Skills group together various field objectives to accomplish the performance requirements in response to highlevel commands. Our framework calculates trajectories that comply with priority constraints while optimizing the desired task objectives in their null spaces. We use a model-based dynamics approach that provides a direct map from field objectives to vehicle accelerations, yielding smooth and accurate trajectory generation. We develop a real-time software system that implements the proposed methods and simulates the coordinated behaviors in a 3D graphical environment. To validate the methodology, we simulate a large exploration task and demonstrate that we can effectively enforce the required constraints while optimizing the exploration goals. I. INTRODUCTIONThe development of technologies for the surveillance, monitoring, and gathering of information in large geographical areas is an important research domain for assessing and planning complex cooperative scenarios. An important problem in this context is the coordination of multitudes of robotic vehicles under network connective and geographical constraints.One of the main challenges in this domain is the execution of many low level objectives as part of the global coordination strategy. This problem arises in our application, in which we must impose network topology and collision avoidance constraints while simultaneously executing path tracking objectives and formation behaviors. Another challenge we are faced with is the design of an architecture that scales efficiently to very large numbers of vehicles.To address these challenges, we employ potential field criteria extensively. The main advantages of using potentials is the low computational overhead associated with task representations and the simplicity of using gradient descent controllers. We develop a centralized system, which allows us to guarantee network connectivity, facilitate decision making at the group level, and create vehicle formations based on global criteria. An important characteristic of our methods is its hierarchical structure, which provides a layer to analyze and resolve possible conflicts between task objectives.To implement potential field strategies, we create a generalized dynamic model of the robotic group that relates vehicle accelerations to control fields. This model provides an effective interface to project artificial potentials into actuator space. The