GMNLP at SemEval-2023 Task 12: Sentiment Analysis with Phylogeny-Based Adapters
This work addresses sentiment analysis for low-resource African languages, presenting an incremental improvement through adapter tuning and ensembling.
The paper tackled sentiment analysis for African languages in the SemEval-2023 AfriSenti-SemEval task, achieving the best F1-score on the Amharic track with a 6.2-point improvement over the second-best system and ranking 5th overall among 10 systems across 15 tracks.
This report describes GMU's sentiment analysis system for the SemEval-2023 shared task AfriSenti-SemEval. We participated in all three sub-tasks: Monolingual, Multilingual, and Zero-Shot. Our approach uses models initialized with AfroXLMR-large, a pre-trained multilingual language model trained on African languages and fine-tuned correspondingly. We also introduce augmented training data along with original training data. Alongside finetuning, we perform phylogeny-based adapter tuning to create several models and ensemble the best models for the final submission. Our system achieves the best F1-score on track 5: Amharic, with 6.2 points higher F1-score than the second-best performing system on this track. Overall, our system ranks 5th among the 10 systems participating in all 15 tracks.