who's currently working on selectional preferences in the context of natural language generation? i'm specifically interested in generation of arguments given predicate, rather than the other way around. (and not interested in probing unless it involves a generation component.)
Conversation
something like InFillmore (aclanthology.org/2021.starsem-1) is close, but i'm trying to find something ontology-free.
1
Replying to
You might also be interested in our work on Tailor (arxiv.org/pdf/2107.07150), which guides generation with control codes that derive from PropBank representations
Show additional replies, including those that may contain offensive content
Show