Quang Huu Pham

Hoang Viet Trinh, Tung Tien Bui, Tam Minh Nguyen et al.

The overwhelming abundance of data has created a misinformation crisis. Unverified sensationalism that is designed to grab the readers' short attention span, when crafted with malice, has caused irreparable damage to our society's structure. As a result, determining the reliability of an article has become a crucial task. After various ablation studies, we propose a multi-input model that can effectively leverage both tabular metadata and post content for the task. Applying state-of-the-art finetuning techniques for the pretrained component and training strategies for our complete model, we have achieved a 0.9462 ROC-score on the VLSP private test set.

Quang Huu Pham, Viet Anh Nguyen, Linh Bao Doan et al.

Natural language processing is a fast-growing field of artificial intelligence. Since the Transformer was introduced by Google in 2017, a large number of language models such as BERT, GPT, and ELMo have been inspired by this architecture. These models were trained on huge datasets and achieved state-of-the-art results on natural language understanding. However, fine-tuning a pre-trained language model on much smaller datasets for downstream tasks requires a carefully-designed pipeline to mitigate problems of the datasets such as lack of training data and imbalanced data. In this paper, we propose a pipeline to adapt the general-purpose RoBERTa language model to a specific text classification task: Vietnamese Hate Speech Detection. We first tune the PhoBERT on our dataset by re-training the model on the Masked Language Model task; then, we employ its encoder for text classification. In order to preserve pre-trained weights while learning new feature representations, we further utilize different training techniques: layer freezing, block-wise learning rate, and label smoothing. Our experiments proved that our proposed pipeline boosts the performance significantly, achieving a new state-of-the-art on Vietnamese Hate Speech Detection campaign with 0.7221 F1 score.