Do Entailment Models know about Reasoning Temporal Ordering on Clinical Texts?

• Autores: Edgar Andres Santamaria, Oier López de Lacalle Lecuona , Aitziber Atutxa Salazar , Koldobika Gojenola Galletebeitia
• Localización: Procesamiento del lenguaje natural, ISSN 1135-5948, Nº. 74, 2025, págs. 349-362
• Idioma: inglés
• Títulos paralelos:
  • ¿Saben razonar los modelos de vinculación sobre el orden temporal en textos clínicos?
• Enlaces
  • Texto completo
• Resumen
  • español

    Este artículo investiga el uso de métodos basados en Inferencia del Lenguaje Natural (NLI) para la extracción de relaciones temporales en textos clínicos (CTRE), abordando desafíos como la escasez de datos, el desequilibrio de las etiquetas y la complejidad específica del dominio. Al reformular la tarea como un problema de NLI, el enfoque reduce los requisitos de anotación y mejora la generalización entre conjuntos de datos. Los experimentos con los corpus THYME y E3C muestran que los modelos basados en NLI superan a los clasificadores tradicionales en entornos de bajos recursos, lo que demuestra una fuerte transferibilidad y robustez frente al desequilibrio de clases, lo que los convierte en una solución eficaz para CTRE en narrativas clínicas.

  • English

    This paper investigates the use of entailment-based methods for Clinical Temporal Relation Extraction (CTRE), addressing challenges such as data scarcity, label imbalance, and domain-specific complexity. By reframing the task as a Natural Language Inference (NLI) problem, the approach reduces annotation requirements and improves generalization across datasets. Experiments with the THYME and E3C corpora show that NLI-based models outperform traditional classifiers in lowresource settings, demonstrating strong transferability and resilience to class imbalance, making them an effective solution for CTRE in clinical narratives.

• Referencias bibliográficas
  • Alfattni, G., N. Peek, and G. Nenadic. 2020. Extraction of temporal relations from clinical free text: A systematic review of current approaches....
  • Allen, J. F. 1983. Maintaining knowledge about temporal intervals. Communications of the ACM, 26(11):832–843.
  • Baldini Soares, L., N. FitzGerald, J. Ling, and T. Kwiatkowski. 2019. Matching the blanks: Distributional similarity for relation learning....
  • Baucells, I., B. Calvo, M. Villegas, and O. L. de Lacalle. 2023. Entailment-based task transfer for catalan text classification in small data...
  • Bowman, S. R., G. Angeli, C. Potts, and C. D. Manning. 2015. A large annotated corpus for learning natural language inference. In L. Márquez,...
  • Bui, A. A. T., D. R. Aberle, and H. Kangarloo. 2007. Timeline: Visualizing integrated patient records. IEEE Transactions on Information Technology...
  • Caron, A., E. Chazard, J. Muller, R. Perichon, L. Ferret, V. Koutkias, R. Beuscart, J.-B. Beuscart, and G. Ficheur. 2017. Itcares: an interactive...
  • Dagan, I., O. Glickman, and B. Magnini. 2006. The pascal recognising textual entailment challenge. In J. Quiñonero-Candela, I. Dagan, B. Magnini,...
  • Dalianis, H. 2018. Clinical text mining: Secondary use of electronic patient records. Springer Nature.
  • Devlin, J., M.-W. Chang, K. Lee, and K. Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding....
  • Hirsch, J., J. Tanenbaum, S. Gorman, C. Liu, E. Schmitz, D. Hashorva, A. Ervits, D. Vawdrey, M. Sturm, and N. Elhadad. 2014. Harvest, a longitudinal...
  • Johnson, A. E. W., M. M. Ghassemi, S. Nemati, K. E. Niehaus, D. A. Clifton, and G. D. Clifford. 2016. Machine learning and decision support...
  • Knez, T. and S. Zitnik. 2024. Multimodal learning for temporal relation extraction in clinical texts. Journal of the American Medical Informatics...
  • Lin, C., T. Miller, D. Dligach, S. Bethard, and G. Savova. 2019. A BERT-based universal model for both within- and crosssentence clinical...
  • Lin, C., T. Miller, D. Dligach, S. Bethard, and G. Savova. 2021a. EntityBERT: Entity-centric masking strategy for model pretraining for the...
  • Lin, C., T. Miller, D. Dligach, S. Bethard, and G. Savova. 2021b. EntityBERT: Entity-centric masking strategy for model pretraining for the...
  • Lin, Y., K. Lu, S. Yu, T. Cai, and M. Zitnik. 2023. Multimodal learning on graphs for disease relation extraction. Journal of Biomedical Informatics,...
  • Liu, Y., M. Ott, N. Goyal, J. Du, M. Joshi, D. Chen, O. Levy, M. Lewis, L. Zettlemoyer, and V. Stoyanov. 2019. Roberta: A robustly optimized...
  • Magnini, B., B. Altuna, A. Lavelli, M. Speranza, and R. Zanoli. 2020. The e3c project: Collection and annotation of a multilingual corpus...
  • Magnini, B., B. Altuna, A. Lavelli, M. Speranza, and R. Zanoli. 2021. The e3c project: European clinical case corpus. Language, 1(L2):L3.
  • Magnini, B., B. Altuna, A. Lavelli, M. Speranza, and R. Zanoli. 2022. E3c annotation guidelines. Accessed: 2022-03-04.
  • Nie, Y., A. Williams, E. Dinan, M. Bansal, J. Weston, and D. Kiela. 2020. Adversarial NLI: A new benchmark for natural language understanding....
  • Ning, Q., Z. Feng, and D. Roth. 2017. A structured learning approach to temporal relation extraction. In Proceedings of the 2017 Conference...
  • Ning, Q., S. Subramanian, and D. Roth. 2019. An improved neural baseline for temporal relation extraction. In K. Inui, J. Jiang, V. Ng, and...
  • Obamuyide, A. and A. Vlachos. 2018. Zero-shot relation classification as textual entailment. In J. Thorne, A. Vlachos, O. Cocarascu, C. Christodoulopoulos,...
  • Olex, A. L. and B. T. McInnes. 2021. Review of temporal reasoning in the clinical domain for timeline extraction: Where we are and where we...
  • Poliak, A., A. Haldar, R. Rudinger, J. E. Hu, E. Pavlick, A. S. White, and B. Van Durme. 2018. Collecting diverse natural language inference...
  • Sainz, O., I. Gonzalez-Dios, O. Lopez de Lacalle, B. Min, and E. Agirre. 2022. Textual entailment for event argument extraction: Zero- and...
  • Sainz, O., O. Lopez de Lacalle, G. Labaka, A. Barrena, and E. Agirre. 2021. Label verbalization and entailment for effective zero and few-shot...
  • Soares, L. B., N. Fitzgerald, J. Ling, and T. Kwiatkowski. 2019. Matching the blanks: Distributional similarity for relation learning. arXiv...
  • Styler IV, W. F., S. Bethard, S. Finan, M. Palmer, S. Pradhan, P. C. de Groen, B. Erickson, T. Miller, C. Lin, G. Savova, and J. Pustejovsky....
  • Thorne, J., A. Vlachos, C. Christodoulopoulos, and A. Mittal. 2018. FEVER: a large-scale dataset for fact extraction and VERification. In...
  • Uppal, S., V. Gupta, A. Swaminathan, H. Zhang, D. Mahata, R. Gosangi, R. R. Shah, and A. Stent. 2020. Two-step classification using recasted...
  • van der Linden, S., J. J. van Wijk, and M. Funk. 2021. Multiple scale visualization of electronic health records to support finding medical...
  • Vashishtha, S., A. Poliak, Y. K. Lal, B. Van Durme, and A. S. White. 2020. Temporal reasoning in natural language inference. In T. Cohn, Y....
  • Vaswani, A., N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin. 2017. Attention is all you need. In...
  • Wang, J., X. Le, X. Peng, and C. Chen. 2023. Adaptive hinge balance loss for document-level relation extraction. In H. Bouamor, J. Pino, and...
  • Wang, Q., T. Mazor, T. A. Harbig, E. Cerami, and N. Gehlenborg. 2022. Threadstates: State-based visual analysis of disease progression. IEEE...
  • Williams, A., N. Nangia, and S. Bowman. 2018. A broad-coverage challenge corpus for sentence understanding through inference. In M. Walker,...
  • Wright-Bettner, K. and M. Palmer. Synthesizes and expands on the thyme annotation guidelines, clinical coreference annotation guidelines,...
  • Zhou, W. and M. Chen. 2022. An improved baseline for sentence-level relation extraction. In Y. He, H. Ji, S. Li, Y. Liu, and C.-H. Chang,...
  • Zhou, Y., Y. Yan, R. Han, J. H. Caufield, K.-W. Chang, Y. Sun, P. Ping, and W.Wang. 2021. Clinical temporal relation extraction with probabilistic...