Model Card for SocRoBERTa-social

Model Description

Based on this paper, this is the SocRoBERTa-social language model. A language model that is trained to better classify social texts in the ESG domain.

Note: We generally recommend choosing the SocialBERT-social model since it is quicker, less resource-intensive and only marginally worse in performance.

Using the SocRoBERTa-base model as a starting point, the SocRoBERTa-social Language Model is additionally fine-trained on a 2k social dataset to detect social text samples.

How to Get Started With the Model

See these tutorials on Medium for a guide on model usage, large-scale analysis, and fine-tuning.

You can use the model with a pipeline for text classification:

from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline

tokenizer_name = "ESGBERT/SocRoBERTa-social"
model_name = "ESGBERT/SocRoBERTa-social"
 
model = AutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(tokenizer_name, max_len=512)
 
pipe = pipeline("text-classification", model=model, tokenizer=tokenizer) # set device=0 to use GPU
 
# See https://huggingface.co/docs/transformers/main_classes/pipelines#transformers.pipeline
print(pipe("We follow rigorous supplier checks to prevent slavery and ensure workers' rights.", padding=True, truncation=True))

More details can be found in the paper

@article{Schimanski23ESGBERT,
    title={{Bridiging the Gap in ESG Measurement: Using NLP to Quantify Environmental, Social, and Governance Communication}},
    author={Tobias Schimanski and Andrin Reding and Nico Reding and Julia Bingler and Mathias Kraus and Markus Leippold},
    year={2023},
    journal={Available on SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4622514},
}