Update README.md

README.md

@@ -1,6 +1,7 @@
-Please refer to [climatebert/econbert](https://huggingface.co/climatebert/econbert) for more information about the model and cite the corresponding paper if you use this model.
+This repository contains a version of [climatebert/econbert](https://huggingface.co/climatebert/econbert) with a corrected folder structure, ensuring compatibility with standard Hugging Face Transformers methods (AutoTokenizer, AutoModel, AutoModelForSequenceClassification).
+
+For more information about the original model, please refer to the official repository [climatebert/econbert](https://huggingface.co/climatebert/econbert), and cite the corresponding [paper](https://dx.doi.org/10.2139/ssrn.5263616) if you use this model.
 
-This repo contains all the necessary tokenizer/model files with a corrected folder structure that enables easier downstream usage such as fine-tuning for sentiment classification.
 
 ## Usage
 
@@ -18,4 +19,6 @@ from transformers import AutoModelForSequenceClassification
 model = AutoModelForSequenceClassification.from_pretrained(model_name, num_labels=3)
 ```
 
-Note that the standard transformers library loading methods (AutoTokenizer, AutoModel) seem to fail with the original repository due to non-standard file organization.
+I do not recommend fine-tuning this model for sentiment classification on the replication data of [Nițoi et al. (2023)](https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/40JFEK) as its performance is not competitive with fine-tuned BERT-base models. For this task, fine-tuned BERT-based models, as proposed by Nițoi et al. (2023), are available here:\
+https://huggingface.co/brjoey/CBSI-bert-large-uncased \
+https://huggingface.co/brjoey/CBSI-bert-base-uncased