Five different input formats are supported:
- Text in surface form (e.g., A baby is playing with a dog)
- Part-of-speech tagged lemmas (e.g., baby#n be#v play#n dog#n). We support only four parts of speech: nouns (n), verbs (v), adjectives (a), and adverbs (r).
- WordNet 3.0 synset offsets (e.g., 09827683-n, 01072949-v, 02084071-n)
- WordNet 3.0 sense keys (e.g., baby%1:18:00::, play%2:33:00::, or dog%1:05:00::)
- Word senses according to WordNet 3.0 sense order (e.g., baby.n.1 play.v.1 dog.n.1 or baby#n#1 play#v#1 dog#n#1)
