Facebook has announced the release of BlenderBot 3, the first 175B-parameter, publicly available chatbot complete with model weights, code, datasets, and model cards.
BlenderBot 3 is capable of searching the internet to chat about virtually any topic, and it’s designed to learn how to improve its skills and safety through natural conversations and feedback from people “in the wild.” Most previous publicly available datasets are typically collected through research studies with annotators that can’t reflect the diversity of the real world.
Meta AI combined two recently developed machine learning techniques, SeeKeR and Director, to build conversational models that learn from interactions and feedback. A focal point of their research is to ensure appropriate safety measures for the chatbot during this process. They developed new techniques that enable learning from helpful teachers while avoiding learning from people who are trying to trick the model into unhelpful or toxic responses.
Initial experiments already show that as more people interact with the model, the more it learns from its experiences and the better and safer it becomes over time — though safety remains an open problem.
The company says it is committed to sharing organic conversational data collected from the interactive demo system as well as model snapshots in the future. They hope this work will help the wider AI community spur progress in building ever-improving intelligent AI systems that can interact with people in safe and helpful ways.
A live interactive conversational AI demo is available here.