| C86-1138 | parser does not waste time in . | hypothesizing | local details that can not possibly |
| E97-1014 | the algorithmic procedure for | hypothesizing | discourse segments based on evidence |
| D09-1138 | suggests that we were reasonable in | hypothesizing | that distributionally similar |
| D08-1060 | function ) words . Instead of | hypothesizing | spontaneous words based only |
| E93-1027 | advance ( by linguists ) , though | hypothesizing | lacking knowledge may require |
| E91-1018 | respect to the spelling rules ( and | hypothesizing | nulls where appropriate ) . On |
| E93-1027 | algorithm and executed before | hypothesizing | any rule or any lexical entry |
| C90-2070 | the role of world knowledge in | hypothesizing | the meaning of " treed " in expressions |
| D09-1140 | proach . As stated earlier , we are | hypothesizing | that the increasing levels of |
| A92-1005 | , a top-down step consists in | hypothesizing | fillers and verifying the presence |
| E93-1033 | unexpected response by a speaker Si by | hypothesizing | that 52 has mistaken some speech |
| C94-2134 | adjuncts ill t10UU phrases , the | hypothesizing | program generates a new rewriting |
| D09-1140 | multilevel abstraction that we are | hypothesizing | will be useful for learning new |
| C94-2134 | successfifl categories in rule | hypothesizing | . The bomnlary condition is given |
| A00-2005 | equal weight . That is , correctly | hypothesizing | a constituent in the reference |
| E93-1027 | a sentence , the grammar rule | hypothesizing | program ( shortly GRHP ) investigates |
| E93-1027 | 4 Generation of Hypotheses 4.1 | Hypothesizing | Grammar Rules from Parsing Failures |
| A92-1005 | words that can support them . | Hypothesizing | headers from already parsed fillers |
| D14-1216 | indirectly touch upon this property by | hypothesizing | that non-compositional idiomatic |
| D15-1182 | we select it for comparison , | hypothesizing | it may have better calibration |
