Viewing tag Long-Distance Dependencies
Number of test suites
8
Average performance
76.27%
Description
No description has been provided for this tag.
Test suites tagged with Long-Distance Dependencies
-
Cleft Structure
- The pseudo-cleft construction involves (i) an extraction of a targeted constituent from a sentence and (ii) a constituent that provides the semantic contents of the targeted constituent and must match it in syntactic category, where (i) and (ii) are linked by the copula. The pseudo-cleft construction can target both NPs and VPs; in the latter case, the VP of the free relative becomes an inflected form of "do". This means that a free relative plus the copula can set up a requirement for the syntactic category that comes next. If the free relative clause has a "do" VP without a direct object, then the main-clause postcopular predicate can be a VP. Otherwise, the postcopular predicate must be an NP.
-
Cleft Structure (with modifier)
- The pseudo-cleft construction involves (i) an extraction of a targeted constituent from a sentence and (ii) a constituent that provides the semantic contents of the targeted constituent and must match it in syntactic category, where (i) and (ii) are linked by the copula. The pseudo-cleft construction can target both NPs and VPs; in the latter case, the VP of the free relative becomes an inflected form of "do". This means that a free relative plus the copula can set up a requirement for the syntactic category that comes next. If the free relative clause has a "do" VP without a direct object, then the main-clause postcopular predicate can be a VP. Otherwise, the postcopular predicate must be an NP.
-
Filler-Gap Dependencies (3 sentential embeddings)
- A filler-gap dependency can be created by extracting any of a number of elements from a subordinate clause, including subjects, objects, indicrect objects (argument structural NPs) and NPs nested in a prepositional phrase. These possibilities serve as the basis for several test suites on filler-gap dependencies. We require that a model successfully pass a two-part criterion for each item in the filler-gap dependency tests: the wh-filler (such as "who" or "what") should make the unextracted filled argument structure more surprising when they exist, and should make the post-gap material less surprising when the argument structure NPs are gapped.
-
Filler-Gap Dependencies (4 sentential embeddings)
- A filler-gap dependency can be created by extracting any of a number of elements from a subordinate clause, including subjects, objects, indicrect objects (argument structural NPs) and NPs nested in a prepositional phrase. These possibilities serve as the basis for several test suites on filler-gap dependencies. We require that a model successfully pass a two-part criterion for each item in the filler-gap dependency tests: the wh-filler (such as "who" or "what") should make the unextracted filled argument structure more surprising when they exist, and should make the post-gap material less surprising when the argument structure NPs are gapped.
-
Filler-Gap Dependencies (hierarchy)
- A filler-gap dependency can be created by extracting any of a number of elements from a subordinate clause, including subjects, objects, indicrect objects (argument structural NPs) and NPs nested in a prepositional phrase. These possibilities serve as the basis for several test suites on filler-gap dependencies. We require that a model successfully pass a two-part criterion for each item in the filler-gap dependency tests: the wh-filler (such as "who" or "what") should make the unextracted filled argument structure more surprising when they exist, and should make the post-gap material less surprising when the argument structure NPs are gapped.
-
Filler-Gap Dependencies (object extraction)
- A filler-gap dependency can be created by extracting any of a number of elements from a subordinate clause, including subjects, objects, indicrect objects (argument structural NPs) and NPs nested in a prepositional phrase. These possibilities serve as the basis for several test suites on filler-gap dependencies. We require that a model successfully pass a two-part criterion for each item in the filler-gap dependency tests: the wh-filler (such as "who" or "what") should make the unextracted filled argument structure more surprising when they exist, and should make the post-gap material less surprising when the argument structure NPs are gapped.
-
Filler-Gap Dependencies (extraction from prepositional phrase)
- A filler-gap dependency can be created by extracting any of a number of elements from a subordinate clause, including subjects, objects, indicrect objects (argument structural NPs) and NPs nested in a prepositional phrase. These possibilities serve as the basis for several test suites on filler-gap dependencies. We require that a model successfully pass a two-part criterion for each item in the filler-gap dependency tests: the wh-filler (such as "who" or "what") should make the unextracted filled argument structure more surprising when they exist, and should make the post-gap material less surprising when the argument structure NPs are gapped.
-
Filler-Gap Dependencies (subject extraction)
- A filler-gap dependency can be created by extracting any of a number of elements from a subordinate clause, including subjects, objects, indicrect objects (argument structural NPs) and NPs nested in a prepositional phrase. These possibilities serve as the basis for several test suites on filler-gap dependencies. We require that a model successfully pass a two-part criterion for each item in the filler-gap dependency tests: the wh-filler (such as "who" or "what") should make the unextracted filled argument structure more surprising when they exist, and should make the post-gap material less surprising when the argument structure NPs are gapped.
Results for Long-Distance Dependencies
| Test suite | Average accuracy |
|---|---|
| Test suite | Average accuracy |
| Cleft Structure | 89.38% |
| Cleft Structure (with modifier) | 65.00% |
| Filler-Gap Dependencies (3 sentential embeddings) | 76.87% |
| Filler-Gap Dependencies (4 sentential embeddings) | 64.97% |
| Filler-Gap Dependencies (hierarchy) | 75.00% |
| Filler-Gap Dependencies (object extraction) | 84.90% |
| Filler-Gap Dependencies (extraction from prepositional phrase) | 68.23% |
| Filler-Gap Dependencies (subject extraction) | 85.42% |