BERTGEN: Multi-task Generation Through BERT

Mitzalis Faidon, Caglayan Ozan, Madhyastha Pranava, Specia Lucia. Arxiv 2021

[Paper]    
Applications BERT Ethics And Bias Model Architecture Multimodal Models

We present BERTGEN, a novel generative, decoder-only model which extends BERT by fusing multimodal and multilingual pretrained models VL-BERT and M-BERT, respectively. BERTGEN is auto-regressively trained for language generation tasks, namely image captioning, machine translation and multimodal machine translation, under a multitask setting. With a comprehensive set of evaluations, we show that BERTGEN outperforms many strong baselines across the tasks explored. We also show BERTGEN’s ability for zero-shot language generation, where it exhibits competitive performance to supervised counterparts. Finally, we conduct ablation studies which demonstrate that BERTGEN substantially benefits from multi-tasking and effectively transfers relevant inductive biases from the pre-trained models.

Similar Work