This paper describes the system and results of our team participated in SemEval-2020 Task4: Commonsense Validation and Explanation (ComVE), which aim to distinguish meaningful natural language statements from unreasonable natural language statements. This task contains three subtasks: Subtask A–Validation, Subtask B–Explanation (Multi-Choice), and Subtask C– Explanation (Generation). In these three subtasks, we only participated in Subtask A, which aims to distinguish whether a given two natural language statements with similar wording are meaningful. To solve this problem, we proposed a method using a combination of BERT with the Bidirectional Gated Recurrent Unit (Bi-GRU). Our model achieved an accuracy of 0.836 in Subtask A (ranked 27/45).

This paper describes the system and results of our team’s participation in SemEval-2020 Task5: Modelling Causal Reasoning in Language: Detecting Counterfactuals, which aims to simulate counterfactual semantics and reasoning in natural language. This task contains two subtasks: Subtask1–Detecting counterfactual statements and Subtask2–Detecting antecedent and consequence. We only participated in Subtask1, aiming to determine whether a given sentence is counterfactual. In order to solve this task, we proposed a system based on Ordered Neurons LSTM (ON-LSTM) with Hierarchical Attention Network (HAN) and used Pooling operation for dimensionality reduction. Finally, we used the K-fold approach as the ensemble method. Our model achieved an F1 score of 0.7040 in Subtask1 (Ranked 16/27).

This article describes the system submitted to SemEval-2020 Task 12 OffensEval 2: Multilingual Offensive Language Recognition in Social Media. The task is to classify offensive language in social media. The shared task contains five languages (English, Greek, Arabic, Danish, and Turkish) and three subtasks. We only participated in subtask A of English to identify offensive language. To solve this task, we proposed a system based on a Bidirectional Gated Recurrent Unit (Bi-GRU) with a Capsule model. Finally, we used the K-fold approach for ensemble. Our model achieved a Macro-average F1 score of 0.90969 (ranked 27/85) in subtask A.