The page lists 25 deliverables related to the research project "QT21".
title and desprition | type | last update |
---|---|---|
Final Report: Morphologically rich languagesThis deliverable will summarize all approaches that have been investigated in this project to improve translation quality on tasks involving morphologically rich languages. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-05-30 |
Improved Learning for Machine Translation (Final vn)This Deliverable describes the results of the Task T1.3, in the form of papers published at major conferences complemented by a summary of those results. It will be available at M18 as a preliminary report, and updated at M36 (final version). Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Semantics in Shallow Models (Final vn)This Deliverable describes the results of the Task T1.2, in the form of papers published at major conferences complemented by a summary of those results. It will be available at M18 as a preliminary report, and updated at M36 (final version). Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Human-informed Continuous Learning: Report 2: Leveraging post-editing + Year 1 annotationsContinuous learning will focus on the three levels (upstream, MT system, downstream) in coordination with the other two WP3 tasks. Research and reporting timing will be synchronised with the availability to the project of annotated data (T3.1) and diagnostic tools (T3.2). M24: upstream/MT/downstream-level components leveraging post editions and Year1 error annotations. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Human-informed Continuous Learning: Report 3: Leveraging post-editing + Year 2 annotationsContinuous learning will focus on the three levels (upstream, MT system, downstream) in coordination with the other two WP3 tasks. Research and reporting timing will be synchronised with the availability to the project of annotated data (T3.1) and diagnostic tools (T3.2). M36: upstream/MT/downstream-level components leveraging post editions and Year2 error annotations. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Data Collection: Resources and ToolsCollected data, and description of data collection procedure in terms of protocols and guidelines for human annotators and interface functionalities of the tools needed for the collection will be presented in this deliverable. The amount and type of information and annotation acquired will be reported for each language pair involved in WP3. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Other | 2019-03-11 |
Data Management Plan Update (Final vn)D5.7 Data Management Plan Update (DFKI, M30) Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Intermediate Report: Under-resourced languagesThis report will summarize the efforts to improve state-of-the-art translation systems for under-resourced languages. It will describe the approaches to learn translation rules from other data sources than parallel data. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Report on the Third Quality Translation Shared TaskReport on the Third Quality Translation Shared Task (UEDIN, M36) Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Public Progress Report (M19-36)Publishable summary of project results and impact during the second period. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Report on the Second Quality Translation Shared TaskReport on the Second Quality Translation Shared Task (UEDIN, M24) Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Final Report: Under-resourced languagesThis report will summarize our work on improving translation quality for under-resources languages. It will contain models to improve learning on other knowledge sources as well as improving the usage of small amounts of available parallel data. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Quality Estimation Metrics and Analysis of Second Annotation Round and Error ProfilesThis deliverable will evaluate the results of MT output from improved systems, comparing it to the results from D3.3, and the progress on fine-grained quality estimation models. It will also report on error profiles identified in the source and target languages, along with a recommendation for how to implement these in an automatic fashion. The report will address the significance of the factors identified and provide hypotheses for future testing in additional languages not covered in QT21. It will include an updated comprehensive database with annotated segments from both rounds of annotation. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Semantic Translation Models (Final vn)This Deliverable describes the results of the Task T1.1, in the form of papers published at major conferences complemented by a summary of those results. It will be available at M18 as a preliminary report, and updated at M36 (final version). Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Evaluation Metrics and Analysis of First Annotation RoundThis deliverable will report on the results of the first annotation round, including error profiles for each language pair and a comparison of the results with analysis based on post-edited examples. It will also include a database of the annotated data for use in training systems and for further analysis and a listing and analysis of measurable factors (semantics and linguistics analyses) found in the annotation done in T3.1 that correspond to human quality judgments. This deliverable will also evaluate the progress on syntax and semantic-informed evaluation metrics. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-11 |
Human-informed Continuous Learning: Report 1: Leveraging post-editingContinuous learning will focus on the three levels (upstream, MT system, downstream) in coordination with the other two WP3 tasks. Research and reporting timing will be synchronised with the availability to the project of annotated data (T3.1) and diagnostic tools (T3.2). M12: downstream-level component leveraging post editions. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-01 |
Semantic Translation ModelsThis Deliverable describes the results of the Task T1.1, in the form of papers published at major conferences complemented by a summary of those results. It will be available at M18 as a preliminary report, and updated at M36 (final version). Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-01 |
Intermediate Report: Morphologically rich languagesThis deliverable will report first experiments on modelling words in morphologically rich languages. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-01 |
Public Progress Report (M1-18))Publishable summary of project results and impact during the first period. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-01 |
Semantics in Shallow ModelsThis Deliverable describes the results of the Task T1.2, in the form of papers published at major conferences complemented by a summary of those results. It will be available at M18 as a preliminary report, and updated at M36 (final version). Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-01 |
Harmonised Error MetricThe harmonised metric will consist of a union of the MQM error type hierarchy and the DQF analytic metric. The harmonisation will involve the addition of necessary categories from DQF to MQM and the adoption of shared terminology. TAUS will determine which categories (if any) to add to DQF from the MQM master hierarchy. The result will be a joint publication of the results and use of the harmonised terminology for error types in both MQM and DQF. The harmonised metric will be tested on a set of approximately 300 segments of EN>DE data taken from T3.1 with triple annotation to ensure that annotators are able to deploy the metric correctly and to suggest improvements to training materials. Testing will be conducted in cooperation with GALA. The result will be a report on agreement and suggestion for any needed changes to the metric profile chosen to be used in subsequent tasks. Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-01 |
Report on the First Quality Translation Shared TaskReport on the First Quality Translation Shared Task (UEDIN, M12) Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-01 |
Data Management Plan UpdateD5.6 Data Management Plan Update (DFKI, M18) Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-01 |
Improved Learning for Machine TranslationThis Deliverable describes the results of the Task T1.3, in the form of papers published at major conferences complemented by a summary of those results. It will be available at M18 as a preliminary report, and updated at M36 (final version). Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-03-01 |
Data Management PlanD5.5 Data Management Plan (DFKI, M06) Programme: H2020-EU.2.1.1.4. - Topic(s): ICT-17-2014 |
Documents, reports | 2019-02-12 |