Text Classification
Transformers
Safetensors
English
bert
AI
Artificial-Intellegence
AI-Disclosure
Finance
BERT
Financial-NLP
Sentence-Classification
Transformers
Banking
Machine-Learning
text-embeddings-inference
Instructions to use bilalzafar/BankAI-BERT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use bilalzafar/BankAI-BERT with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="bilalzafar/BankAI-BERT")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("bilalzafar/BankAI-BERT") model = AutoModelForSequenceClassification.from_pretrained("bilalzafar/BankAI-BERT", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Update README.md
Verified signature54dffe8 -
Update README.md
Verified signature8abf7a0 -
Update README.md
Verified signature6e4b5f3
-
🗑️ remove Supplementary Material.zip
Verified signature085f6f3 -
Update README.md
Verified signature73b49a7
-
Update README.md
Verified signature8218575 -
Update README.md
Verified signature1154177 -
Update README.md
Verified signaturec626747 -
Update README.md
Verified signature6147802
-
Supplementary Material
Verified signature2f72c7e -
🗑️ remove Supplementary Material.zip
Verified signature26e0f65 -
Update README.md
Verified signaturee509bd8 -
Update README.md
Verified signature0379ceb -
Update README.md
Verified signaturee3fcbb2 -
Update README.md
Verified signaturedf34946 -
Update README.md
Verified signature8bf0f42 -
Update README.md
Verified signaturee8b3cab -
Update README.md
Verified signature0898eac -
Update README.md
Verified signatureaf4ecb2 -
Supplementary Material
Verified signature1c0586c -
FinAI-BERT Model File
Verified signaturedb72167 -
🗑️ remove tokenizer_config.json
Verified signature4cbb7cb -
🗑️ remove special_tokens_map.json
Verified signature7229ca3 -
🗑️ remove config.json
Verified signature45924b0 -
🗑️ remove Supplementary material.zip
Verified signaturedd051cb -
🗑️ remove vocab.txt
Verified signature3e5ef66 -
🗑️ remove tokenizer.json
Verified signature4b651d0 -
🗑️ remove model.safetensors
Verified signature549fb23
-
Update README.md
Verified signatured628dac -
Update README.md
Verified signature21dedb7
-
Update README.md
Verified signature6ae321c -
Update README.md
Verified signature5973504 -
Update README.md
Verified signature4e390d4 -
Update README.md
Verified signature61b2ffb -
Create README.md
Verified signature# FinAI-BERT FinAI-BERT is a domain-specific BERT-based model fine-tuned for detecting AI-related disclosures in financial texts, particularly from banking annual reports. It was trained using a balanced dataset of 1,586 sentences (793 AI, 793 Non-AI) using `bert-base-uncased` as the base model. ## Intended Use FinAI-BERT is designed to assist researchers, analysts, and regulators in identifying AI narratives in financial disclosures at the sentence level. ## Performance - Accuracy: 99.7% - F1-score: 0.997 - ROC AUC: 1.000 - Brier Score: 0.0000 ## Training Data The model was trained on a manually annotated dataset from U.S. bank reports (2015–2023), covering AI and non-AI sentences. ## Files Included - `config.json`, `tokenizer.json`, `vocab.txt`, `model.safetensors`: Model files - `tokenizer_config.json`, `special_tokens_map.json`: Tokenizer configuration ## Usage ```python from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_…88e95e5 -
Supplementary material
Verified signatureThe supplementary material is provided to support the implementation and replication of FinAI-BERT.06dfaac -
Delete README.md
Verified signature2bfe3c0 -
Update README.md
Verified signature# FinAI-BERT **FinAI-BERT** is a domain-specific BERT-based model fine-tuned for detecting AI-related disclosures in financial texts, particularly banking annual reports. It was trained using a balanced dataset of 1,586 sentences (793 AI, 793 Non-AI) based on `bert-base-uncased`. ## Intended Use Designed for use by researchers, analysts, and regulators to identify AI-related discourse at the sentence level in financial documents. ## Performance - **Accuracy**: 99.7% - **F1-score**: 0.997 - **ROC AUC**: 1.000 - **Brier Score**: 0.0000 ## Training Details Trained on a manually labeled dataset from US banking reports (2015–2023). The model is robust in identifying AI narratives and offers explainability via SHAP. ## Files Included - `model.safetensors`, `config.json`, `tokenizer.json`, `vocab.txt`: Model files - `tokenizer_config.json`, `special_tokens_map.json`: Tokenizer settings ## Citation > Zafar, B. et al. (2025). *Detecting Artificial Intelligence Disclosures in Financial Tex…0b7d493 -
FinAI-BERT pretrained model files
Verified signaturec6f62a0
-
initial commit
Verified signature7836742