Twenty Representative Cases with Pathway Extraction Analysis
This supplementary document presents twenty representative cases of pathway extraction analysis in Transformer models. It provides additional experimental visualizations and case-level results to complement the analyses presented in the associated doctoral dissertation. The document is intended to support the interpretation and examination of the extracted pathways and their relationships to target-specific model behavior.
Authors
- Min Xue (ORCID: https://orcid.org/0000-0001-6577-1082)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-10-30
- DOI
- https://doi.org/10.5281/zenodo.23804317
- Primary Topic
- Explainable Artificial Intelligence (XAI)
- Type
- article
- Field-Weighted Citation Impact
- 0.00