BART-10K-Summarization

This model is a fine-tuned version of Facebook's BART model specifically for summarizing financial 10K report sections.

Model description

BART-10K-Summarization is designed to produce concise summaries of detailed financial reports, assisting analysts and stakeholders in quickly understanding key information without needing to parse the entire document.

Intended uses & limitations

This model is intended to aid financial analysts, investors, and regulatory bodies by summarizing sections of 10K reports. It may not perform well on non-financial texts or highly technical documents outside the scope of standard financial reporting.

Training and evaluation data

The model was trained on a curated dataset of 10K financial reports, each annotated with executive summaries by experienced financial analysts.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3.0

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.1+cu121
• Tokenizers 0.19.1