English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Children acquire their native language with apparent ease by observing howlanguage is used in context and attempting to use it themselves. They do sowithout laborious annotations, negative examples, or even direct corrections.We take a step toward robots that can do the same by training a groundedsemantic parser, which discovers latent linguistic representations that can beused for the execution of natural-language commands. In particular, we focus onthe difficult domain of commands with a temporal aspect, whose semantics wecapture with Linear Temporal Logic, LTL. Our parser is trained with pairs ofsentences and executions as well as an executor. At training time, the parserhypothesizes a meaning representation for the input as a formula in LTL. Threecompeting pressures allow the parser to discover meaning from language. First,any hypothesized meaning for a sentence must be permissive enough to reflectall the annotated execution trajectories. Second, the executor -- a pretrainedend-to-end LTL planner -- must find that the observe trajectories are likelyexecutions of the meaning. Finally, a generator, which reconstructs theoriginal input, encourages the model to find representations that conserveknowledge about the command. Together these ensure that the meaning is neithertoo general nor too specific. Our model generalizes well, being able to parseand execute both machine-generated and human-generated commands, withnear-equal accuracy, despite the fact that the human-generated sentences aremuch more varied and complex with an open lexicon. The approach presented hereis not specific to LTL: it can be applied to any domain where sentence meaningscan be hypothesized and an executor can verify these meanings, thus opening thedoor to many applications for robotic agents.

Learning a natural-language to LTL executable semantic parser for grounded robotics