Hi @danflick , I am playing again with the SICK dataset [1] of text entailment. I processed all sentences with ERG, asking for at most ten analyses. The following sentences are the only ones that ERG didn't parse. Can you take a look?
A hole is being burrowed by the badger
A prey is quickly running behind the cheetah
The orange colored dices are being cast by a cook to win the pepper
The man is cooking slices of potato
The lady is penciling on eyeshadow
The woman is penciling on eyeshadow
The woman is not penciling on eyeshadow
A person is jotting something with a pencil
A man is drawing on a digital dry erase board
A man is not drawing on a digital dry erase board
A yellow flower is being clung to by a bee
A hurdle is being leapt by a horse that has a rider on its back
A topless woman is being smeared with brown substance and a nobody is in the background
A boy is wearing all black and is sliding down a blow up yellow slide
A boy is wearing all black and is sliding down a blow up slide
A girl is wearing all black and is sliding down a blow up yellow slide
A cyclist is biking in a snowy forest at day
Reference:
Marelli, Marco, et al. "A SICK cure for the evaluation of compositional distributional semantic models." LREC. 2014.
arademaker
commented
1 year ago
Hi @danflick , I am playing again with the SICK dataset [1] of text entailment. I processed all sentences with ERG, asking for at most ten analyses. The following sentences are the only ones that ERG didn't parse. Can you take a look?
Reference: