TariffBERT

Summary

<!-- Provide a quick summary of what the model is/does. -->

TariffBERT is a fine-tuned version of ProsusAI/finbert for financial sentiment analysis focused on tariff and trade-policy news.
It classifies English-language text into Positive, Negative or Neutral sentiment toward tariff-related market impact.

Model Details

• Developed by: Cristobal Medina Meza (@CristobalMe)
• Model type: BERT-based sequence classifier
• Language: English
• License: MIT
• Finetuned from: ProsusAI/finbert

<!-- ### Model Sources [optional] TO DO <!-- Provide the basic links for the model. --> <!-- - **Repository:** [More Information Needed] --> <!-- - **Paper [optional]:** [More Information Needed] --> <!-- - **Demo [optional]:** [More Information Needed] -->

Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

Direct Use

<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->

• Sentiment classification of news articles and headlines, regulatory filings, or analyst notes discussing tariffs, trade wars, or import/export policy.
• Can be used as-is via the Hugging Face pipeline("text-classification").

Downstream Use

<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->

• As a component in financial forecasting, event-driven trading strategies, or risk dashboards.
• Further fine-tuning on sector-specific trade data.

Out-of-Scope Use

<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->

• Non-financial general sentiment tasks (movie reviews, product opinions).
• High-stakes decision-making (e.g., compliance enforcement) without human oversight.

Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

• Domain bias: Training data is tariff/trade news; performance may degrade on unrelated finance text.

• Temporal drift: Model reflects market language up to its training cutoff SEPTEMBER 2025; newer policy jargon may be misclassified.

• Geographic bias: Data may over-represent US trade discourse.

Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

Use confidence thresholds and human review in production.

How to Get Started with the Model

Use the code below to get started with the model.

from transformers import pipeline

pipe = pipeline("text-classification", model="CristobalMe/TariffBERT")

text = "This is an example text for classification."

result = pipe(text)

print(result)

<!-- TO DO ## Training Details TO DO ### Training Data <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> <!-- [More Information Needed] ### Training Procedure <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> <!-- #### Preprocessing [optional] [More Information Needed] #### Training Hyperparameters - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision --> <!-- #### Speeds, Sizes, Times [optional] <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. --> <!-- [More Information Needed] --> <!-- ## Evaluation <!-- This section describes the evaluation protocols and provides the results. --> <!-- ### Testing Data, Factors & Metrics #### Testing Data <!-- This should link to a Dataset Card if possible. -->

Metrics

Accuracy

Results

Accuracy: 0.9

Environmental Impact

<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->

• Hardware Type: Apple MacBook Pro 14″ (M4 Pro, 14-core CPU / 20-core GPU)
• Training Time: ~15 minutes
• Energy Use Estimate: ≈0.02 kWh
• Estimated Carbon Emissions: ≈0.01 kg CO2eq

<!-- ## Citation [optional] <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> <!-- **BibTeX:** [More Information Needed] **APA:** [More Information Needed] -->

Model Card Contact

For questions or collaboration, email [email protected]

Or contact @CristobalMe