RooseBERT-scr-uncased

RooseBERT is a domain-specific BERT-based language model pre-trained on English political debates and parliamentary speeches. It is designed to capture the distinctive features of political discourse, including domain-specific terminology, implicit argumentation, and strategic communication patterns.

This variant — scr-uncased — was trained from scratch (SCR) with a custom uncased WordPiece tokenizer built from the political debate corpus, giving it a vocabulary better suited to political language than standard BERT. The uncased variant lowercases all input text before tokenization, making it more robust to capitalisation variation across debate transcripts.

📄 Paper: RooseBERT: A New Deal For Political Language Modelling
💻 GitHub: https://github.com/deborahdore/RooseBERT

Model Details

The SCR approach trains BERT entirely from scratch on domain-specific data, using a custom tokenizer. This allows political terms like deterrent, endorse, bureaucrat, statutorily, and consequential to be represented as single tokens — whereas standard BERT would split them into multiple sub-tokens. The custom vocabulary shares only ~56% of its tokens with bert-base-uncased.

Training Data

RooseBERT was pre-trained on 11GB of English political debate transcripts spanning 1919–2025, drawn from:

All datasets were sourced from authoritative, official political settings. Pre-processing removed hyperlinks, markup tags, and collapsed whitespace.

Intended Use

RooseBERT is intended as a base model for fine-tuning on downstream NLP tasks related to political discourse analysis. It is especially well-suited for:

• Sentiment Analysis of parliamentary speeches and debates
• Stance Detection (support/oppose classification)
• Argument Component Detection and Classification (claims and premises)
• Argument Relation Prediction and Classification (support/attack/no-relation)
• Motion Policy Classification
• Named Entity Recognition in political texts

The uncased variant is recommended when capitalisation is inconsistent across your data, or when your task does not rely on case distinctions (e.g., proper nouns vs. common nouns).

How to Use

from transformers import AutoTokenizer, AutoModelForMaskedLM

tokenizer = AutoTokenizer.from_pretrained("ddore14/RooseBERT-scr-uncased")
model = AutoModelForMaskedLM.from_pretrained("ddore14/RooseBERT-scr-uncased")

For fine-tuning on a downstream classification task:

from transformers import AutoTokenizer, AutoModelForSequenceClassification

tokenizer = AutoTokenizer.from_pretrained("ddore14/RooseBERT-scr-uncased")
model = AutoModelForSequenceClassification.from_pretrained(
    "ddore14/RooseBERT-scr-uncased",
    num_labels=2
)

# Recommended fine-tuning hyperparameters (from paper):
# learning_rate ∈ {2e-5, 3e-5, 5e-5}
# batch_size ∈ {8, 16, 32}
# epochs ∈ {2, 3, 4}

Note: This model uses an uncased tokenizer. Make sure not to pass do_lower_case=False when loading — lowercasing is applied automatically.

Evaluation Results

RooseBERT was evaluated across 10 datasets covering 6 downstream tasks. Results below are for RooseBERT-scr-uncased (Macro F1 unless noted).

RooseBERT-scr-uncased outperforms BERT-base-uncased on 8 out of 10 tasks. Results are averaged over 5 runs with different random seeds. For tasks where capitalisation carries semantic weight (e.g., proper noun-heavy NER), the cased variant may perform better.

Perplexity on held-out political debate data:

Available Variants

SCR (from scratch) models use a custom political vocabulary; CONT (continued pre-training) models are initialised from the original BERT weights and vocabulary. Cased models preserve capitalisation; uncased models lowercase all input.

Limitations

• RooseBERT is trained exclusively on English political debates. Cross-lingual use is not supported.
• The model may reflect biases present in official political speech, including over-representation of certain geopolitical perspectives.
• Performance on NER tasks does not benefit from domain-specific pre-training when entity categories are general rather than politically specific.
• As an uncased model, it loses information from capitalisation, which may matter for tasks involving proper nouns or acronyms.
• As with all encoder-only models, RooseBERT is best suited to classification and labelling tasks rather than generation.

Citation

If you use RooseBERT in your research, please cite:

@article{dore2025roosebert,
  title={RooseBERT: A New Deal For Political Language Modelling},
  author={Dore, Deborah and Cabrio, Elena and Villata, Serena},
  journal={arXiv preprint arXiv:2508.03250},
  year={2025}
}

Acknowledgements

This work was supported by the French government through the 3IA Côte d'Azur programme (ANR-23-IACL-0001). Computing resources were provided by GENCI at IDRIS (grant 2026-AD011016047R1) on the Jean Zay supercomputer.