<ul><li>Post hoc explanation methods have emerged to enhance model transparency by attributing model outputs to input features, but face challenges due to specificity to certain neural network architectures and data modalities.</li><li>Existing XAI frameworks have limitations in flexibility to diverse model architectures and data modalities, restricted number of supported XAI methods, and sub-optimal recommendations of explanations.</li><li>To address these limitations, PnPXAI is introduced as a universal XAI framework supporting diverse data modalities and neural network models in a Plug-and-Play manner.</li><li>PnPXAI automatically detects model architectures, recommends applicable explanation methods, and optimizes hyperparameters for optimal explanations, showcasing its effectiveness across various domains like medicine and finance.</li></ul>

PnPXAI: A Universal XAI Framework Providing Automatic Explanations Across Diverse Modalities and Models

Discover more