We present BlenderBot 3, a 175B parameter dialogue model capable of open-domain conversation with access to the internet and a longterm memory, and having been trained on a large number of user defined tasks. We release both the model weights and code, and have also deployed the model on a public web page to interact with organic users. This technical report describes how the model was built (architecture, model and training scheme), and details of its deployment, including safety mechanisms. Human evaluations show its superiority to existing open-domain dialogue agents, including its predecessors Komeili et al., 2022). Finally, we detail our plan for continual learning using the data collected from deployment, which will also be publicly released. The goal of this research program is thus to enable the community to study ever-improving responsible agents that learn through interaction. * * We use the phrase continual learning in the sense of learning that continues over time using data from the model's interactions, but training itself will actually be performed in successive large batches; the model is not updated online.† Equal contribution.
Object recognition and manipulation are critical for enabling robots to operate within a household environment. There are many grasp planners that can estimate grasps based on object shape, but these approaches often perform poorly because they miss key information about non-visual object characteristics, such as weight distribution, fragility of materials, and usability characteristics. Object model databases can account for this information, but existing methods for constructing 3D object recognition databases are time and resource intensive, often requiring specialized equipment, and are therefore dicult to apply to robots in the eld. We present an easy-to-use system for constructing object models for 3D object recognition and manipulation made possible by advances in web robotics. The database consists of point clouds generated using a novel iterative point cloud registration algorithm, which includes the encoding of manipulation data and usability characteristics. The system requires no additional equipment other than the robot itself, and non-expert users can demonstrate grasps through an intuitive web interface with virtually no training required. We validate the system with data collected from both a crowdsourcing user study and a set of grasps demonstrated by an expert user. We show that the crowdsourced grasps can produce successful autonomous grasps, and furthermore the demonstration approach outperforms purely vision-based grasp planning approaches for a wide variety of object classes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.