ErgSemantics_Raw_Materials

This is a page for us to keep links to questions on the delphinqa site, ERG issue tracker, etc that could help further populate the ErgSemantics pages. This might include nice answers elaborated on one of those fora, but also just questions that should be answered even if not yet.

• https://delphinqa.ling.washington.edu/t/converting-mrs-output-to-a-logical-form/413

The need here is for the reference docs to include some description of what they are implying in terms of logical evaluation. I'll use all the wrong terminology here but I hope my point comes through: For example, _car_n_1(x) logically means "restrict x to those nouns that are a car". neg(eh) logically means...I'm not even sure how to explain it, but there must be a logical model we have in mind for determining the truth condition of the predicate since we wrote it. I really think we need to explain it. _which_q(xhh) the two scopal arguments share a "logical evaluation" model with many predicates, but I suspect there is something special going on. And I know there is for d_number_q(xhh) because Dan described it here. It would be extraordinarily helpful if we described the logical model that is being used for different predicates. Especially those with scopal args.

• https://delphinqa.ling.washington.edu/t/interpretation-of-on-p-state-for-drop-the-diamond-on-the-table/483

as per Dan in that thread "I don’t see the description we should have on the ERS pages explaining the difference between stative and directional preposition senses"

• https://delphinqa.ling.washington.edu/t/how-to-think-about-mrs-structures-that-use-a-non-verb-as-the-mrs-top-like-where-are-you/491

MRS constructs where something like "blue" is treated as the verb (i.e. index?) of the phrase.

• https://delphinqa.ling.washington.edu/t/what-is-the-discourse-predicate-trying-to-say/548

Need a discussion of the discourse predication.

• https://delphinqa.ling.washington.edu/t/how-to-consistently-get-a-parse-of-quoted-phrases-that-uses-fw-seq-with-quoted-predicates/553 I know fw_seq and friends have been updated by Dan since that post, but I suspect there is a valuable discussion about when they get used, if there is always an fw_seq interpretation from ACE if something is quoted (I hope there is!), etc.