blocks world

A classic AI test domain consisting of blocks (sometimes with pyramids and boxes) resting on a table or on one another, plus a hand or robot arm that can pick up and put down one clear block at a time. Because states and actions can be described exactly in logic, it became the standard illustration for automated planningautomated planning, n.: The branch of AI that works out a sequence of actions to reach a goal, such as the steps a robot must take to assemble a part, from descriptions of what each action needs and what it changes.: STRIPSSTRIPS, prop. n.: An early AI planning program, and the simple action language it introduced, in which each action lists what must be true beforehand, what becomes true and what stops being true.-style actions with preconditions and effects were routinely demonstrated on it, and the Sussman anomaly, a three-block goal that defeats planners which solve subgoals one at a time, exposed the problem of interacting goals. It was also used in early computer visioncomputer vision, n.: The part of AI that lets computers make sense of images and video: recognizing faces, reading signs, spotting tumors on scans or judging how far away a car is. (recovering 3-D polyhedra from line drawings), in language understanding through SHRDLUSHRDLU, prop. n.: A program from around 1970 that let people move toy blocks in a simulated world by typing ordinary English commands and questions, and that could explain what it had done and why., and in physical robot experiments. Blocks-world problems remain standard planning benchmarks and reappear as tests of large language modellarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. planning, where performance can collapse once block names and actions are disguised.

The model solved the standard blocks-world tasks but failed once the blocks were renamed with nonsense words.