| | --- |
| | license: apache-2.0 |
| | --- |
| | FinRoberta is a specialised version of the Roberta Large Model designed for financial classification tasks. FinRoberta has been developed by finetuning the Roberta Large model on a vast corpus of Reuters and DPA news articles, SEC filings, followed by further finetuning on the [Headlines](https://www.kaggle.com/datasets/ankurzing/sentiment-analysis-for-financial-news) dataset to make it capable of handling classification tasks. |
| | ## How to get started with the model |
| |
|
| | Use the code below to get started with the model. |
| | ```python |
| | from transformers import AutoTokenizer, RobertaForSequenceClassification |
| | |
| | tokenizer = AutoTokenizer.from_pretrained('roberta-large') |
| | |
| | model = RobertaForSequenceClassification.from_pretarined('Chaitanya14/finroberta') |
| | ``` |
| | ## Citation |
| | If you use the FinRoberta model, please cite with the following BibTex entry: |
| | ``` |
| | @misc{sinha2025finbloomknowledgegroundinglarge, |
| | title={FinBloom: Knowledge Grounding Large Language Model with Real-time Financial Data}, |
| | author={Ankur Sinha and Chaitanya Agarwal and Pekka Malo}, |
| | year={2025}, |
| | eprint={2502.18471}, |
| | archivePrefix={arXiv}, |
| | primaryClass={cs.IR}, |
| | url={https://arxiv.org/abs/2502.18471}, |
| | } |
| | ``` |
