The Machine Translation Research Unit, FBK, Trento, Italy is looking for two candidates strongly motivated to work on innovative and interdisciplinary projects at the intersection of language and translation technology for a 6-month internship starting in mid-April.
Most likely, due to the lock-downs and other restrictions and conditions related to the COVID-19 pandemic, the internship will be conducted remotely. Nevertheless, the intern will work in strict collaboration with senior researchers and Ph.D. students of the Unit. To ease the smooth progress of the project, regular meetings will take place on a weekly basis. In case of need, additional ones will be quickly organized upon request. *Internship#1. Gender Bias and Automatic Translation* *Description:* Automatic translation systems have been shown to be affected by significant gender bias, especially when translating from natural gender languages (eg. English) into languages with grammatical gender (eg. Italian, French, Spanish). As such, they tend to exacerbate stereotypical gender association in translation, favor masculine over feminine forms in translation and exhibit reduced performance with women speakers. As an interdisciplinary and relatively recent field of inquiry, the extent of the technical, social, and linguistic implications is yet to be fully disclosed. This project aims to study the role that different components have in the generation/exacerbation of gender bias as well as to implement mitigating strategies that alleviate the problem. Requirements: - Ongoing/Completed Academic background in Computer Science/Computational Linguistics degree - Good knowledge of written and spoken English - Basic knowledge of neural networks, Python - (Optional) Knowledge of PyTorch - (Optional) Basic knowledge of Italian, French, Spanish *Application*: https://forms.gle/cqs6YRgFL5sqUs8S8 *Contacts*: Marco Gaido (mga...@fbk.eu), Luisa Bentivogli (bent...@fbk.eu) *Internship#2. Data-to-text Generation* *Description:* The candidate will study and develop advanced Data-to-Text solutions capable of “translating” a record of data (attribute-value pairs, instead of plain text) into text in a given language. Research on this challenging task will be carried out within the framework of an ongoing project involving both local and international partners. Requirements: - Ongoing/Completed Academic background in computer science or similar - Good knowledge of written and spoken English - Good knowledge of Tensorflow, Python, neural networks, and sequence-to-sequence modeling - A background machine translation is a plus *Application*: https://forms.gle/c4oNhHqx2PyuHT3JA *Contacts*: Marco Turchi (tur...@fbk.eu), Matteo Negri (ne...@fbk.eu)
_______________________________________________ Mt-list site list Mt-list@eamt.org http://lists.eamt.org/mailman/listinfo/mt-list
