Restoring, placing and dating ancient texts through collaboration between AI and historians.Read More
Learning Robust Real-Time Cultural Transmission without Human Data
In this work, we use deep reinforcement learning to generate artificial agents capable of test-time cultural transmission. Once trained, our agents can infer and recall navigational knowledge demonstrated by experts. This knowledge transfer happens in real time and generalises across a vast space of previously unseen tasks.Read More
Learning Robust Real-Time Cultural Transmission without Human Data
In this work, we use deep reinforcement learning to generate artificial agents capable of test-time cultural transmission. Once trained, our agents can infer and recall navigational knowledge demonstrated by experts. This knowledge transfer happens in real time and generalises across a vast space of previously unseen tasks.Read More
Probing Image-Language Transformers for Verb Understanding
Multimodal Image-Language transformers have achieved impressive results on a variety of tasks that rely on fine-tuning (e.g., visual question answering and image retrieval). We are interested in shedding light on the quality of their pretrained representations–in particular, if these models can distinguish verbs or they only use the nouns in a given sentence. To do so, we collect a dataset of image-sentence pairs consisting of 447 verbs that are either visual or commonly found in the pretraining data (i.e., the Conceptual Captions dataset). We use this dataset to evaluate the pretrained models in a zero-shot way.Read More
Probing Image-Language Transformers for Verb Understanding
Multimodal Image-Language transformers have achieved impressive results on a variety of tasks that rely on fine-tuning (e.g., visual question answering and image retrieval). We are interested in shedding light on the quality of their pretrained representations–in particular, if these models can distinguish verbs or they only use the nouns in a given sentence. To do so, we collect a dataset of image-sentence pairs consisting of 447 verbs that are either visual or commonly found in the pretraining data (i.e., the Conceptual Captions dataset). We use this dataset to evaluate the pretrained models in a zero-shot way.Read More
Accelerating fusion science through learned plasma control
Successfully controlling the nuclear fusion plasma in a tokamak with deep reinforcement learningRead More
MuZero’s first step from research into the real world
Collaborating with YouTube to optimise video compression in the open source VP9 codec.Read More
Red Teaming Language Models with Language Models
In our recent paper, we show that it is possible to automatically find inputs that elicit harmful text from language models by generating inputs using language models themselves. Our approach provides one tool for finding harmful model behaviours before users are impacted, though we emphasize that it should be viewed as one component alongside many other techniques that will be needed to find harms and mitigate them once found.Read More