Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2505.10515 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908365597179904 |
|---|---|
| author | Kim, Seongun Kim, Sol A Kim, Geonhyeong Menadjiev, Enver Lee, Chanwoo Chung, Seongwook Kim, Nari Choi, Jaesik |
| author_facet | Kim, Seongun Kim, Sol A Kim, Geonhyeong Menadjiev, Enver Lee, Chanwoo Chung, Seongwook Kim, Nari Choi, Jaesik |
| contents | Recently, post hoc explanation methods have emerged to enhance model transparency by attributing model outputs to input features. However, these methods face challenges due to their specificity to certain neural network architectures and data modalities. Existing explainable artificial intelligence (XAI) frameworks have attempted to address these challenges but suffer from several limitations. These include limited flexibility to diverse model architectures and data modalities due to hard-coded implementations, a restricted number of supported XAI methods because of the requirements for layer-specific operations of attribution methods, and sub-optimal recommendations of explanations due to the lack of evaluation and optimization phases. Consequently, these limitations impede the adoption of XAI technology in real-world applications, making it difficult for practitioners to select the optimal explanation method for their domain. To address these limitations, we introduce \textbf{PnPXAI}, a universal XAI framework that supports diverse data modalities and neural network models in a Plug-and-Play (PnP) manner. PnPXAI automatically detects model architectures, recommends applicable explanation methods, and optimizes hyperparameters for optimal explanations. We validate the framework's effectiveness through user surveys and showcase its versatility across various domains, including medicine and finance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_10515 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PnPXAI: A Universal XAI Framework Providing Automatic Explanations Across Diverse Modalities and Models Kim, Seongun Kim, Sol A Kim, Geonhyeong Menadjiev, Enver Lee, Chanwoo Chung, Seongwook Kim, Nari Choi, Jaesik Machine Learning Artificial Intelligence Recently, post hoc explanation methods have emerged to enhance model transparency by attributing model outputs to input features. However, these methods face challenges due to their specificity to certain neural network architectures and data modalities. Existing explainable artificial intelligence (XAI) frameworks have attempted to address these challenges but suffer from several limitations. These include limited flexibility to diverse model architectures and data modalities due to hard-coded implementations, a restricted number of supported XAI methods because of the requirements for layer-specific operations of attribution methods, and sub-optimal recommendations of explanations due to the lack of evaluation and optimization phases. Consequently, these limitations impede the adoption of XAI technology in real-world applications, making it difficult for practitioners to select the optimal explanation method for their domain. To address these limitations, we introduce \textbf{PnPXAI}, a universal XAI framework that supports diverse data modalities and neural network models in a Plug-and-Play (PnP) manner. PnPXAI automatically detects model architectures, recommends applicable explanation methods, and optimizes hyperparameters for optimal explanations. We validate the framework's effectiveness through user surveys and showcase its versatility across various domains, including medicine and finance. |
| title | PnPXAI: A Universal XAI Framework Providing Automatic Explanations Across Diverse Modalities and Models |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2505.10515 |