| W06-2303 | parsing ) and the latter task | PropBank parsing | be - low . These results have |
| N06-2026 | between two parsing tasks : the | PropBank parsing | task and the PTB parsing task |
| W08-2101 | and Yi and Palmer conclude that | PropBank parsing | is too difficult and suffers |
| W06-2303 | between two parsing tasks : the | PropBank parsing | task and the PTB parsing task |
| N06-2026 | we can perform the more complex | PropBank parsing | task at levels of accuracy comparable |
| W06-2303 | syntactic PTB parsing and semantic | PropBank parsing | is strict enough that an integrated |
| W06-2303 | it was trained on , and compute | PropBank parsing | performance measures on the resulting |
| N06-2026 | which it was trained , and compute | PropBank parsing | performance measures on the resulting |
| N06-2026 | syntactic PTB parsing and semantic | PropBank parsing | is strict enough that an integrated |
| W08-2101 | ours , based on a first step of | PropBank parsing | , are comparable to other methods |
| N06-2026 | confirm our initial hypothe PTB and | PropBank parsing | , compared to the original SSN |
