Hitachi at SemEval-2023 Task 3: Exploring Cross-lingual Multi-task Strategies for Genre and Framing Detection in Online News
This work addresses genre and framing detection in low-resource multilingual news analysis, but it is incremental as it builds on existing methods for a specific competition task.
The paper tackled genre and framing detection in online news across multiple languages by exploring cross-lingual and multi-task strategies with pretrained language models, achieving the highest macro-averaged F1 scores in Italian and Russian genre categorization subtasks.
This paper explains the participation of team Hitachi to SemEval-2023 Task 3 "Detecting the genre, the framing, and the persuasion techniques in online news in a multi-lingual setup.'' Based on the multilingual, multi-task nature of the task and the low-resource setting, we investigated different cross-lingual and multi-task strategies for training the pretrained language models. Through extensive experiments, we found that (a) cross-lingual/multi-task training, and (b) collecting an external balanced dataset, can benefit the genre and framing detection. We constructed ensemble models from the results and achieved the highest macro-averaged F1 scores in Italian and Russian genre categorization subtasks.