Tag
An Estonian inventor built a wall-climbing robot named Albert that uses AI to spray-paint murals on buildings, covering up to 240m² per day, 100x faster than manual painting.
This research paper investigates how shortcut solutions learned by Transformer models, specifically BERT, impair their ability to perform continual compositional reasoning. It contrasts BERT with ALBERT, finding that ALBERT's recurrent nature offers better inductive bias for continual learning tasks.