DIRT Paraphrase Collection - RTE Users: Difference between revisions

From ACL Wiki
Jump to navigation Jump to search
Celct (talk | contribs)
No edit summary
Celct (talk | contribs)
No edit summary
Line 8: Line 8:
! nowrap="nowrap"|Specific usage description
! nowrap="nowrap"|Specific usage description
! nowrap="nowrap"|Evaluations / Comments
! nowrap="nowrap"|Evaluations / Comments
|- bgcolor="#ECECEC" align="left"
|- bgcolor="#ECECEC" align="left"
| UIUC
| BIU
| RTE3
| RTE5
|  
|  
| Paired verb/argument patterns
| We used the canonical DIRT rulebase version of Szpektor and Dagan (RANLP 2007), and considered top 25 rules.
| Ablation test performed. Positive impact of the resource: +1.33% accuracy on two-way task.
 
|- bgcolor="#ECECEC" align="left"
| BIU
| RTE4
|  
|  
| We used the canonical DIRT rulebase version of Szpektor and Dagan (RANLP 2007), and considered top 25 rules.
| +0.9% on RTE-4 ablation tests
|- bgcolor="#ECECEC" align="left"
|- bgcolor="#ECECEC" align="left"
| Boeing
| Boeing
Line 22: Line 31:
boeing run1: 67%/6%;<br>
boeing run1: 67%/6%;<br>
boeing run2: 54%/30%
boeing run2: 54%/30%
|- bgcolor="#ECECEC" align="left"
|- bgcolor="#ECECEC" align="left"
| BIU
| UAIC
| RTE4
| RTE4
|  
|  
| We used the canonical DIRT rulebase version of Szpektor and Dagan (RANLP 2007), and considered top 25 rules.
|  
| +0.9% on RTE-4 ablation tests
| ''Data taken from the RTE4 proceedings. Participants are recommended to add further information.''
 
|- bgcolor="#ECECEC" align="left"
|- bgcolor="#ECECEC" align="left"
| UAIC
| UIUC
| RTE4
| RTE3
|  
|  
| Paired verb/argument patterns
|  
|  
| ''Data taken from the RTE4 proceedings. Participants are recommended to add further information.''
 
|- bgcolor="#ECECEC" align="left"
|- bgcolor="#ECECEC" align="left"
| Uoeltg
| Uoeltg
Line 40: Line 52:
|  
|  
| ''Data taken from the RTE4 proceedings. Participants are recommended to add further information.''
| ''Data taken from the RTE4 proceedings. Participants are recommended to add further information.''
|- bgcolor="#ECECEC" align="left"
|- bgcolor="#ECECEC" align="left"
| ''New user''
| ''New user''
Line 46: Line 59:
|  
|  
| ''Participants are encouraged to contribute.''
| ''Participants are encouraged to contribute.''
|}
|}
{|class="wikitable" cellpadding="3" cellspacing="0" border="0" style="margin-left: 20px;"
{|class="wikitable" cellpadding="3" cellspacing="0" border="0" style="margin-left: 20px;"

Revision as of 15:00, 26 November 2009

When not otherwise specified, the data about version, usage and evaluation of the resource have been provided by participants themselves.

Participants* Campaign Version Specific usage description Evaluations / Comments
BIU RTE5 We used the canonical DIRT rulebase version of Szpektor and Dagan (RANLP 2007), and considered top 25 rules. Ablation test performed. Positive impact of the resource: +1.33% accuracy on two-way task.
BIU RTE4 We used the canonical DIRT rulebase version of Szpektor and Dagan (RANLP 2007), and considered top 25 rules. +0.9% on RTE-4 ablation tests
Boeing RTE4 Original DIRT db Elaborate T sentence with DIRT-implied entailments precision/recall in RTE4:

boeing run1: 67%/6%;
boeing run2: 54%/30%

UAIC RTE4 Data taken from the RTE4 proceedings. Participants are recommended to add further information.
UIUC RTE3 Paired verb/argument patterns
Uoeltg RTE4 Data taken from the RTE4 proceedings. Participants are recommended to add further information.
New user Participants are encouraged to contribute.
Total: 5


[*] For further information about participants, click here: RTE Challenges - Data about participants

   Return to RTE Knowledge Resources