Language-Table Robot Language Label Trajectory Dataset

The Language-Table dataset was proposed by Google in 2022 to build robots that can interact in real time through natural language instructions in the real world.Interactive Language: Talking to Robots in Real Time". The dataset contains nearly 600,000 language-labeled trajectories to promote the development of more advanced, capable robots that can interact with natural language. By training on a dataset containing hundreds of thousands of language-annotated trajectories, the researchers found that the resulting policy can execute 10 times more instructions than before, which describe end-to-end visual, auditory and motor skills in the real world. The dataset contains nearly 600,000 language-labeled trajectories to promote the development of more advanced, capable robots that can interact with natural language.