In the "Introduction" section, the authors point out the neurophysiological evidences that a human brain has a hierarchical structure for the object recognition, and that the learning and recognition of objects occurs concurrently in a human brain. Then, they briefly explained how they built their 3D object recognition model that concurrently learns and recognizes the objects that works in the similar way to the human brain. The authors talk about conventional approaches for the object recognition problem in the "Related work" section. They mention about the probabilistic latent semantic indexing (pLSI), latent Dirichlet allocation (LDA) and its variations, etc. And they point out that none of these conventional models can learn and recognize objects in an open-ended manner.

In recent years there has been a tremendous surge in the general capabilities of AI systems, mainly fuelled by training foundation models on internetscale data. Nevertheless, the creation of openended, ever self-improving AI remains elusive. In this position paper, we argue that the ingredients are now in place to achieve openendedness in AI systems with respect to a human observer. Furthermore, we claim that such open-endedness is an essential property of any artificial superhuman intelligence (ASI). We begin by providing a concrete formal definition of open-endedness through the lens of novelty and learnability. We then illustrate a path towards ASI via open-ended systems built on top of foundation models, capable of making novel, humanrelevant discoveries. We conclude by examining the safety implications of generally-capable openended AI. We expect that open-ended foundation models will prove to be an increasingly fertile and safety-critical area of research in the near future.