Abstract -This paper presents our participation in the FinNLP-2023 shared task on multi-lingual en vironmental, social, and corporate governance issue identification (ML-ESG). The task’s ob jective is to classify news articles based on the 35 ESG key issues defined by the MSCI ESG rating guidelines. Our approach focuses on the English and French subtasks, employing the CerebrasGPT, OPT, and Pythia models, along with the zero-shot and GPT3Mix Augmenta tion techniques. We utilize various encoder models, such as RoBERTa, DeBERTa, and Fin BERT, subjecting them to knowledge distilla tion and additional training.
Our approach yielded exceptional results, se curing the first position in the English text sub task with F1-score 0.69 and the second position in the French text subtask with F1-score 0.78. These outcomes underscore the effectiveness of our methodology in identifying ESG issues in news articles across different languages. Our findings contribute to the exploration of ESG topics and highlight the potential of leverag ing advanced language models for ESG issue identification.