Context-aware NMT using Selected Context

An offline context selection technique

This work was presented at an international conference, The author proposed to use pre-trained models and context selection techniques to effectively encode the contextual information. The idea is to select the most significant context from existing data and encode contextual features using pre-trained models to have rich representation of context.

Overview of proposed approach.

There can different context selection methodologies, we can have a number of previous sentences may have some relevant information about the topic and current sentence may have dependency on that. A current sentence that needs to be translated, a number of previous or next sentences can be selected as part of context. Our experiments suggest that considering more than 3 sentences as context may not improve the performance of system.

The graphical illustration of results which indicate selected approaches ourperfromed fixed context baseline.

References

2022

  1. Context-aware neural machine translation using selected context
    Sami Ul Haq, Sadaf Abdul Rauf, Arslan Shaukat, and 1 more author
    In 2022 19th International Bhurban Conference on Applied Sciences and Technology (IBCAST), 2022