Explore projects

Prior work, with the aim of extracting the hidden states from LSTM joey model. Then to used on the DQN approach.

Further experiments on the basis of "Probing the Need for Visual Context in Multimodal Machine Translation" (Caglayan et al, 2019)

Charniuk, Nguyen, Safdel

Code for my Bachelor thesis "Derivative-free Optimization for Sequence-to-Sequence Models".

Kaldi Fork implementing Mozilla Common Voice Training

Neural multi-label Commonsense Relation Classifier as described in Becker at al. 2019

Identification and Resolution of Comparative Anaphora in the English Language

Classify audio recordings of trams into tram type and movement type

Fork of Fairseq to conduct ablation studies for the transformer model