Challenges in Explaining Pretrained Clinical Text Classifiers
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917538426781696 |
|---|---|
| author | Miok, Kristian Klemen, Matej Škrlj, Blaz Šikonja, Marko Robnik |
| author_facet | Miok, Kristian Klemen, Matej Škrlj, Blaz Šikonja, Marko Robnik |
| contents | Explaining the predictions of neural models in clinical NLP remains a significant challenge, especially for complex tasks involving long, unstructured medical texts. While post-hoc methods like LIME and SHAP are widely used, they often fall short when applied to clinical narratives. In this paper, we identify core limitations of token-level and perturbation-based explanation techniques through targeted demonstra- tions on a hospital length-of-stay prediction task. Our findings reveal issues such as overemphasis on non-informative tokens, instability in at- tributions, and high-confidence predictions for incoherent input variants. These results underscore the need for explanation strategies that are clin- ically meaningful, semantically grounded, and robust to linguistic noise. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_28060 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Challenges in Explaining Pretrained Clinical Text Classifiers Miok, Kristian Klemen, Matej Škrlj, Blaz Šikonja, Marko Robnik Computation and Language Explaining the predictions of neural models in clinical NLP remains a significant challenge, especially for complex tasks involving long, unstructured medical texts. While post-hoc methods like LIME and SHAP are widely used, they often fall short when applied to clinical narratives. In this paper, we identify core limitations of token-level and perturbation-based explanation techniques through targeted demonstra- tions on a hospital length-of-stay prediction task. Our findings reveal issues such as overemphasis on non-informative tokens, instability in at- tributions, and high-confidence predictions for incoherent input variants. These results underscore the need for explanation strategies that are clin- ically meaningful, semantically grounded, and robust to linguistic noise. |
| title | Challenges in Explaining Pretrained Clinical Text Classifiers |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2605.28060 |