Learning Neural Sequence-to-Sequence Models from Weak Feedback with Bipolar Ramp Loss

1 code implementation • TACL 2019 • Laura Jehl, Carolin Lawrence, Stefan Riezler

We show that bipolar ramp loss objectives outperform other non-bipolar ramp loss objectives and minimum risk training (MRT) on both weakly supervised tasks, as well as on a supervised machine translation task.

Machine Translation Semantic Parsing +1

Document-Level Information as Side Constraints for Improved Neural Patent Translation

no code implementations • WS 2018 • Laura Jehl, Stefan Riezler

Machine Translation Translation

Learning to translate from graded and negative relevance information

no code implementations • COLING 2016 • Laura Jehl, Stefan Riezler

We present an approach for learning to translate by exploiting cross-lingual link structure in multilingual document collections.

Language Modelling Translation

Source-side Preordering for Translation using Logistic Regression and Depth-first Branch-and-Bound Search

no code implementations • EACL 2014 • Laura Jehl, Adri{\`a} de Gispert, Mark Hopkins, Bill Byrne

Machine Translation regression +1

Boosting Cross-Language Retrieval by Learning Bilingual Phrase Associations from Relevance Rankings

no code implementations • EMNLP 2013 • Artem Sokokov, Laura Jehl, Felix Hieber, Stefan Riezler

Information Retrieval Learning-To-Rank +2

Task Alternation in Parallel Sentence Retrieval for Twitter Translation

no code implementations • ACL 2013 • Felix Hieber, Laura Jehl, Stefan Riezler

Information Retrieval Language Modelling +4

Twitter Translation using Translation-Based Cross-Lingual Retrieval

no code implementations • WS 2012 • Laura Jehl, Felix Hieber, Stefan Riezler

Domain Adaptation Information Retrieval +4