Semantic Role Labeling Quiz

Answers and explanations appear as you take the quiz.

1. What core question does semantic role labeling aim to answer?
   • Who did what to whom, when, where, and why?
   • How positive or negative is the text?
   • What language is the text written in?
   • How many words are in the sentence?
2. In “The book was sold to John by Mary,” which semantic role does Mary fill?
   • The Theme being sold
   • The seller (Agent) of the book
   • The Recipient of the sale
   • A Location where the sale occurred
3. Which resource defines verb-specific argument structures used in SRL?
   • ImageNet
   • WordNet's audio corpus
   • PropBank
   • The Penn Treebank of phone numbers
4. In one common token-level SRL formulation, how are argument spans represented?
   • As image pixels classified by color
   • As audio frames with pitch labels
   • As a single sentiment score for the sentence
   • With BIO-style token labels paired with semantic role labels
5. Why might an SRL system predict the 'sense' of a predicate?
   • Because the same verb can take different argument frames depending on meaning
   • To translate the verb
   • To count syllables
   • To detect the author's mood
6. Which downstream task benefits most directly from SRL identifying a 'when' argument?
   • Spell checking
   • Temporal question answering
   • Font rendering
   • Audio compression
7. What is a predicate-argument structure meant to capture?
   • Only the number of words in a sentence
   • A list of syntactic phrase boundaries with no roles
   • A predicate and the semantic roles of the participants or modifiers associated with it
   • The emotional tone of the author
8. What does PropBank provide for semantic role labeling?
   • A multilingual audio database
   • A collection of image segmentation masks
   • A list of word frequencies without annotations
   • Annotated predicate-argument structures with verb-specific rolesets