dataset_info:
name: Agent or Industry Fraud Harmful
license: CC-BY-4.0
language: en
tags:
- rhesis
- test-sets
- chatbot
- insurance

Dataset Card for Agent or Industry Fraud Harmful

Description

The test set is designed for evaluating the performance of an insurance chatbot in handling various scenarios related to the insurance industry. The focus of the test is to analyze the chatbot's robustness by examining its ability to handle harmful behaviors. Specifically, the test set contains examples that highlight agent or industry fraud. The purpose is to gauge how effectively the chatbot can respond to fraudulent activities and identify potential instances of deceit. Through these evaluations, the test aims to assess the chatbot's reliability and accuracy in detecting and addressing fraud in the insurance domain.

Structure

The dataset includes four key columns:

• Behavior: The performance dimension evaluated (Reliability, Robustness, or Compliance).
• Topic: The topic validated as part of the prompt.
• Category: The category of the insurance-related task, such as claims, customer service, or policy information.
• Prompt: The actual test prompt provided to the chatbot.
• Source: Provides a reference to the source used for guidance while creating the test set.

Disclaimer

Some test cases may contain sensitive, challenging, or potentially upsetting content. These cases are included to ensure thorough and realistic assessments. Users should review test cases carefully and exercise discretion when utilizing them.

Integration

In order to easily integrate a Rhesis test set into your existing development pipeline, you can make use of the Rhesis SDK.

Prerequisites

1. Create an API key by signing up at app.rhesis.ai.

2. Install the Rhesis SDK:

   pip install rhesis-sdk
   

3. Use the following Python snippet to access the dataset:

from rhesis.entities import TestSet

# Initialize and load the test set
test_set = TestSet(id="agent-or-industry-fraud-harmful")
df = test_set.load()  # Returns a pandas DataFrame

# Alternatively, you can download the CSV file directly
test_set.download()  # Downloads to current directory as test_set_{id}.csv

Using the Rhesis SDK, you get access to the entire collection of Rhesis test sets, including this one. You can also create your own test sets.

For further details on how to integrate the SDK into your workflow, refer to the Rhesis SDK documentation.

Community

We welcome contributions and discussions from the community! Here are the different ways you can get involved:

• GitHub: Report issues or contribute to the Rhesis SDK on our GitHub repository.
• Discord: Join our Discord server to connect with other users and developers.
• Email: Reach out to us at [email protected] for support or inquiries.

Sources

The following sources were used in creating this dataset:

• Deng, B., Wang, W., Feng, F., Deng, Y., Wang, Q., & He, X. (2023). Attack prompt generation for red teaming and defending large language models. arXiv preprint arXiv:2310.12505.

Citation

If you use this dataset, please cite:

@inproceedings{rhesis,
  title={Rhesis: A Testbench for Evaluating LLM Applications - Agent or Industry Fraud Harmful},
  author={Rhesis},
  year={2025}
}