INSURBIAS (Insurance Bias Benchmark) is a counterfactual audit benchmark for evaluating gender bias in multimodal large language model (LLM) insurance claim adjudication. It supports the paper:
Huang, Fei, Shamim, Md Mushahidul Islam, Khern-am-nuai, Warut, and Cohen, Maxime C. Beyond the Binary: Gender Bias in LLM-Evaluated Insurance Claims. Available at SSRN:… See the full description on the dataset page: https://huggingface.co/datasets/feihuangfh/INSURBIAS.
